No man is an iland

Email software popularity: 5 lessons for your list

2009-11-05T09:36:00.002+01:00

Who cares what software people use to read your emails?

If you have a "safe" email design, you know each message displays gracefully whether viewed in Outlook 2007 or Gmail.

The only exception is when people use a mobile device, but you can get round that with the assumption that they'll save your mail to view on a desktop later.

But there is value to knowing exactly where your emails are viewed.

Getting the stats

You may be wondering how on earth you can tell whether people are viewing your messages on AOL or Apple Mail. It's only recently that the right tools have become available.

I used the client distribution component of the wider MailboxIQ service to analyze my list. FingerPrint is a standalone service or you may have access to a solution through your ESP (example).

Here's a snapshot summary of the 50 or so different clients and webmail services my list used to view a recent newsletter issue:

83% viewed the email using desktop software (like Outlook), 15% using a webmail service (like Yahoo! Mail) and some 2% using a mobile device.

So what?

1. Compare with benchmarks

The first thing you can do is compare the numbers with benchmarks to see how your list differs and then think about why that might be so:

Email client popularity stats by Campaign Monitor (June 2009)
Email client statistics for B2B and B2C by Fingerprint (Sept 2008)
Recipient platform preferences for B2B and B2C by Pivotal Veracity (Oct 2009)

It seems I have an usually large number of people viewing on Outlook 2007 (even for a B2B list), Apple Mail and Gmail.

This suggests the list is attracting a business audience that's updating its software faster than most, perhaps a little more first-mover, tech-savvy than others and with a significant minority of design/creative individuals?

Supporting evidence for that interpretation comes from the data provided by MailboxIQ on the browsers used to view my messages in webmail environments:

Both Firefox and Chrome are more popular with subscribers than you'd expect given broader stats on browser market share.

Such knowledge might influence my content strategy going forward. It also worries me that my email's design is pretty simplistic: what must those cutting-edge and creative folk think?

2. Compare with your list

Software stats are only recorded when an open/render is recorded (more on that later). Your standard campaign reports should also tell you which subscribers "open" an email.

So you can calculate the percentage of recorded opens associated with a webmail address (i.e. how many of your gmail.com addresses recorded an open) and then compare this with the results from your software/webmail distribution stats.

Let's take the newsletter issue used to produce the software stats above.

7.9% of opens recorded by my ESP were from subscribers with a gmail.com address, 4.2% subscribers with a yahoo.com address and 3.2% subscribers with a hotmail.com address.

All these numbers are higher than the equivalent number produced by MailboxIQ, suggesting that people are signing up with a webmail address but downloading their webmail to another viewing environment.

If the results were reversed, it would suggest many people are signing up with business domain addresses, but actually viewing the mail in a consumer webmail environment: either directly or because their email applications are actually powered by Gmail, etc..

[Gmail recently announced that over 20 million users do this.]

In reality, both these activities are happening. Your stats simply tell you which is happening more often.

The clear message is this: you can't make assumptions about viewing environments based on the domain name of the email address.

B2B marketers may also be surprised at the volume of webmail users on their list (15% in my case), suggesting they need to pay just as much attention to webmail deliverability issues (particularly sender reputation) as their B2C counterparts.

3. Trend spotting and mobile strategies

If you follow your stats through time, you can pick up software trends that perhaps reflect changes in the makeup of your list. Most importantly, this data helps you decide on whether (and how) to tackle the issue of mobile email.

I have a small list and only 2% use mobile devices to view my emails. For now, it makes little sense to develop a fully-fledged mobile email strategy, with mobile-ready landing pages etc. But what if that number was 10%?

4. Design testing

Of course, if you haven't got a "safe" email design, information on subscriber software use lets you know exactly what display environments you should be testing.

Perhaps you have a B2C list and never worried too much about Outlook 2007? Or a B2B list and ignored Windows Live Hotmail? Now you know if you were right to do so.

Even if your design is "safe", there are things to learn. Most design testing tools do not include software/browser combinations. In other words, you get a single screenshot of how your design looks in Gmail. And maybe it looks just great.

But does it look great in Gmail when viewed in IE8, Firefox, Chrome and Safari? For example, I never really bothered to worry about Google's Chrome browser. But now I see 17% of my Gmail users also use Chrome, maybe it's worth investigating.

5. Customer-level design

What if you could associate a particular email software or webmail service with an individual email address? Could you then begin sending emails optimized for that particular display environment?

The possibilities are many. For example,

Including detailed "add to address list" instructions that are a perfect match in terms of vocabulary and instructional steps
Changing subject lines to fit the likely available space (especially for webmail users)
Streamlined versions for mobile users
Dumping inline CSS for an external stylesheet or "CSS in head" approach for those environments that support it (saving bandwidth costs)
etc...

The danger here is that people sometimes switch between software or webmail services. But you could build in rules: if someone records the same software over a certain time period, then you can feel safe sending them future emails customized for that software.

Theoretically you could do this anyway for the webmail domains on your subscriber list. But as we've already seen, the domain name in the email address does not necessarily tell you where the subscriber actually reads their email.

Obviously, you'd need to test to see if creating such customized versions was justified by the results. But the potential is clear, especially when combined with other targeting technologies, such as trigger emails.

A note on measurement issues

As mentioned, a recipient needs to "open" an email for the tools to capture data on the software that recipient is using. So no data is recorded where no open is registered.

Pretty much every major email client or webmail service has image blocking in place, which prevents the open tracking image from displaying.

If we assume that people's propensity to activate images is independent of the software they use, then this technical problem is irrelevant: everyone is equally underrepresented. But there are still issues to take into account when interpreting software distribution numbers:

1. The assumption isn't necessarily true. Recent data from MailChimp, for example, showed that Gmail users tend to engage more with email than other webmail users (i.e. they are more likely to open email). So maybe the stats overestimate Gmail use.

2. Some display environments don't have the facility to display (tracking) images at all. So, for example, certain mobile devices will be heavily underrepresented in the stats.

3. One-off deliverability problems can skew the results. If you trigger a block at Yahoo.com, well, the number of people viewing your email using Yahoo.com is...um...likely to be low.

If you trigger a block at Postini, then corporate users see less of your email than webmail users. The result: a false impression of how many people use software like Outlook.

Keep those issues in mind. In particular, you might want to average numbers over several campaigns so that short-term or one-off delivery problems don't bias the stats too much.

Future of deliverability: 4. The role of certification

2009-10-30T14:39:00.003+01:00

Part 1: User interaction
Part 2: Authentication
Part 3: Domain-based reputation

If your emails are certified by an authoritative third-party, then you get priority delivery treatment at ISPs that recognize that certification.

Sound good?

Yes, no, maybe: the pros and cons of getting your emails certified are outlined here.

In theory, certification simplifies life for ISPs and others managing incoming email by pre-identifying "good" messages.

So some might suggest (I have in the past) that certification is the future of email deliverability.

But does our panel of experts agree?

Will certification become a must have?

First off, it seems very unlikely that email stamps will become a serious proposition, where everyone has to pay to get email delivered.

Jeremy Saibil, Director of Deliverability at Campaigner says:

"I don't think pay-for-delivery is going to happen. Unless there is a radical change in how internet interaction is billed overall...email has purposes other than those associated with marketing, so the needs of personal and professional users will trump paid email."

But will some kind of paid certification become critical for marketing email?

Chris Wheeler, Director of Deliverability at Bronto, says:

"I would never say that 3rd party accreditation and certifications will be a must for every sender. They make financial sense for some senders who need an extra lift in inbox delivery or want the added assurance emails will be automatically rendered, but this won't replace responsible email marketing."

Even those running certification programs agree that not everyone will need it. George Bilbrey (President) and Tom Sather (Professional Services Director) of Return Path say:

"Email is never going to be 100% pay-to-play, which is a good thing. Good email marketers will always have good deliverability."

Does this mean certification is primarily for bad senders?

If low delivery rates come from poor email practices, then is certification a way for bad senders to pay to get into the inbox?

No.

You need to satisfy rigorous standards before obtaining certification: ISPs are not going to give special treatment to certified email if it means more unwanted messages landing in their customers' inbox.

The credibility and growth of certification therefore depends on certifiers maintaining those standards, even if it means turning down paying customers.

Which means senders whose poor email practices cause them delivery problems probably can't get certified. Saibil notes:

"...businesses that would consistently benefit from certification will likely not qualify for it."

But do good senders need certification if they already have good delivery rates?

Conversely, if you run a high-quality email program, then the potential delivery lift through certification may not be big enough to justify the expense. Saibil again:

"...at the end of the day, if your email marketing practices are good, you won't need certification to deliver the mail..."

Deirdre Baird, President & CEO of Pivotal Veracity adds:

"...we probably will never reach a day when all mail is certified because if mailers are engaging in industry-established best practices and have a winning marketing strategy, they don't need a 3rd party to certify their email as deserving of inbox placement: users will do that."

She continues:

"Its also critically important to note that "Certification" or "Safe Listing" is not going to insulate the mailer against poor deliverability."

"...the ISPs' move to customer-level preferences and engagement as the highest-priority filter significantly minimizes the impact these types of programs can have on folder placement."

So who does benefit from certification, if bad senders don't qualify and good senders don't need it?

Well, even small improvements might justify the costs of certification and there are still real delivery benefits. One is what we might call delivery insurance. As Saibil notes:

"Many of our clients use certification as a safety net to guard against content issues or similar surprises that can periodically negatively impact on their delivery rates."

Bilbrey and Sather also say:

"...we are seeing a lot of ISPs shift their focus toward finding ways to identify good email in an attempt to reduce false positives. With this wider acceptance by ISPs we think the benefits for marketers will make certification programs like ours even more attractive and more cost-effective."

...and certification is not just about deliverability

What many forget is that certification has other potential benefits.

For example, participating ISPs may not block images on emails certified by partners. Some certification programs cause icons to be displayed that mark you out as a certified sender, which might bring trust benefits.

Long-term, certification may be linked to other benefits, such as support for more design functionality (e.g. scripts) that current webmail services typically block automatically.

Baird says:

"Certification will continue to hold appeal for some marketers, especially those whose messages are frequently being spoofed and those that need assurance that critical communications are bypassing filters (even then, end-user preferences prevail, by the way)."

Each sender needs to assess each certification alternative in the light of their current situation, the relative benefit to their current performance levels, and the cost of complying with the standards and paying for the certification process.

It seems unlikely that certification will come to dominate deliverability in the way that reputation etc. does.

But certification's role seems set to grow in a wider context, through the overall mix of potential benefits for design, delivery, trust-building etc.

Still to come: a look at B2B lists and reputation, plus a heap of useful links...

More on deliverability

Future of deliverability: 3. The role of domain reputation

2009-10-28T17:25:00.002+01:00

Part 1: User interaction
Part 2: Authentication

When organizations look at incoming email, they use a set of criteria to decide what to do with it.

For many such organizations, particularly the big webmail services, the reputation of the sender is a very important criterion determining whether that email should go to the inbox.

This sender reputation is itself built out of various factors, such as how many spam complaints the sender gets or how many defunct addresses they are trying to email.

It's a pretty good way of regulating email, but problems arise through the definition of the "sender" part of sender reputation.

To date, this reputation has largely been tied not to the sender in the traditional sense of the word, but to the sending IP address: the original "connection" to the net that initiated the email transfer.

This gives rise to various difficulties.

For example, if different organizations send email through the same "connection" (sharing an IP address at an email marketing service, for example) then they also share a common sender reputation.

So a "good" email sender can find their reputation dragged down by the others. And vice versa.

Even if you have a dedicated IP address all to your own, you can face problems if you move to a new one. Your sender reputation isn't portable: you have to start from scratch.

If we could associate reputation with an actual domain (e.g. news.email-marketing-reports.com), then reputation would become independent of the system/location used to send out that domain's emails.

It wouldn't matter where my emails are sent out from, because the reputation factors associated with those emails would be tied to the news.email-marketing-report.com domain name.

I wouldn't have to worry about naughty senders putting out their email through the same "connection" as me: it wouldn't affect my domain-based reputation.

This concept is called domain-based sender reputation and it's already impacting the email deliverability world.

Unfortunately, the reality is not quite as simple and positive as the above theory suggests. Nevertheless, the potential benefits are clear.

Deirdre Baird, President & CEO of Pivotal Veracity told me:

"...domain-based reputation will be of big help in addressing the pain-points associated with sharing IP addresses...furthermore, large mailers will have the benefit of being able to add new IPs and/or switch IPs (or ESPs) without the painful and blocking-fraught 'IP warm up process'."

Domain-based reputation works both ways, of course, as George Bilbrey (President) and Tom Sather (Professional Services Director) of Return Path explain:

"If you are a good mailer, sharing IP space with less-good mailers, you no longer pay the penalty for their practices. But if you are a bad mailer sharing IP space with good mailers, no more free ride."

Is domain-based reputation relevant today?

Understandably, ISPs and others are cagey about providing details, but domain reputation is already a factor at some organizations and likely to spread rapidly.

Baird reveals:

"...many leading providers are already in the process of moving to domain-based reputation. Yahoo is already enabling DKIM-compliant senders to benefit from domain-based reputation portability. AOL is switching later in Q4/early Q1."

Bilbrey and Sather confirm that the impacts are already being felt, noting that the top-tier ESPs they work with have started to see some domains get treated differently off the same IP addresses.

Jeremy Saibil, Director of Deliverability at Campaigner offers further confirmation:

"I believe we'll be seeing domain-based reputation "kick in" almost immediately. In fact, I have suspicions it has already started; certainly "warming up" new DKIM authenticated IPs has been much easier of late with certain ISPs."

These quotes highlight the importance of authentication, particularly DKIM (see Part 2 for details).

Domain reputation only works if the receiver can verify whether the domain claiming to be the sender is indeed the source of the email.

The implication is clear. If you want to benefit from domain-based sender reputation, then authenticate your messages. Equally, since not everyone will do so, "traditional" IP-based reputation will continue to play a role in deliverability (more on that later).

Domain-based reputation is not a "get out of jail free" card

Of course, the advantages of domain-based reputation only accrue if you have a good one (a point often overlooked).

Bilbrey and Sather note that the metrics that make up the current IP reputation systems are going to be the same for domain reputation: complaints, unknown users, spam trap hits, sending consistency, "this is not spam" votes, opens and clicks:

"Bottom line: if you have a bad IP reputation because of poor practices, domain reputation is not going to improve your inbox placement rates."

They also warn that the mix of factors and thresholds for these factors will also likely change as ISPs adapt their approaches:

"What defines a "good" complaint rate, for example, is not static and never will be."

So marketers need to stay on top of new developments.

Will it replace IP-based sender reputation?

A question that's hard to answer is the extent to which domain-based reputation will truly replace the current role played by IP-based reputation.

One thing that's certain: the two will inevitably coexist for the foreseeable future. Chris Wheeler, Director of Deliverability at Bronto, says:

"...IP reputation will still be used by many ISPs while domain reputation gains momentum. Also, IP and domain reputation heuristics will not be mutually exclusive."

Instead, he says, domain-based and IP-based reputation will blend together in determining the final destination of delivered email.

Baird also believes IP-based reputation will remain important:

"...the switch to domain-based reputation is just beginning, and many smaller ISPs haven't implemented DKIM or SPF yet. Additionally, we don't expect IP-based reputation to go away entirely as it is used when no domain-reputation exists even at the largest ISPs."

Bilbrey and Sather add:

"The truth is that ISPs are using both domain and IP reputation. And that is not going to change anytime soon. ISPs will still be looking at reputation metrics for IPs, IP-ranges, URLs and more."

So domain-based reputation isn't a global panacea to the issues surrounding shared IPs. It makes you more accountable for your actions, but you won't become completely independent of other senders' email activities if you share an IP address with them.

Bilbrey and Sather continue:

"A high-performing marketer with a stellar domain reputation that is sharing an IP with bottom-of-the-barrel spammers is still likely to see issues getting to the inbox, even with domain reputation in place at both ends of the pipe."

For the individual marketers, it's important to know if your ESP is using DKIM authentication. If not, as Bilbrey and Sather note, today's rules continue to apply because...

"...in the absence of DKIM, ISPs are going to fall back to IP reputation."

Another issue is exactly how DKIM is implemented. Saibil explains:

"...it will be interesting to see how ESPs roll out DKIM/DK...For shared IP environments that support hundreds of thousands of small customers, will it be feasible for each customer to sign as well as the ESP? Perhaps eventually, but I suspect the majority of those emails will be signed by the ESP only for some time to come."

Should different email types be sent from different domains?

Domain-based sender reputation brings accountability, which is a welcome development if you're a good sender. But even the best senders can run into temporary deliverability problems.

If your domain reputation takes a hit, then all email from that domain might struggle to get to inboxes.

Does it make sense, then, to allocate different message types to different (sub-)domains?

Saibil notes that those using an ESP for marketing email likely already have this domain split:

"Using an ESP generally requires a separate domain entity so the ESP can do reply processing."

...but he adds:

"Splitting corporate, transactional and marketing mails at a domain level is something I recommend, however, we usually suggest using sub-domains for these splits."

Bilbrey and Sather go into further detail:

"There is a string in the DKIM record, d=, which is what reputation is tied to. So you can use different d= strings (examples: d= marketing.returnpath.net, d=sales.returnpath.net, d= service.returnpath.net) to differentiate various streams even though the recipient still just sees "returnpath.net"."

Baird warns, however, that while email authentication can help make the process of separating mail streams more efficient and clearly defined, marketers should not rely on authentication alone.

After all, as we just learned, domain-based reputation isn't the only factor affecting deliverability.

"There should be physical separations between corporate mail and all other mail. If a marketer's mail stream comes under fire and you're blocked, your corporate mail may be the only line of communication to a filtering body/ISP. At the very least it's important to leave yourself a life raft in case the whole ship goes down."

Wheeler also points us to this detailed post on domain-based reputation, where he recommends auditing all your outgoing email to identify any potential conflicts:

"If Bob over in coupon sales is killing it with blasting recipients twice daily and aggravating them, Sally over in order processing trying to get shipment confirmation emails out will suffer if both mail streams are coming from the same domain."

Part 4: Email certification

More on deliverability

Current and future use of email: info sources

2009-10-23T12:35:00.000+02:00

A reader asked me if there was a central repository for all the statistics and articles rebuffing the regular media hype suggesting that "(insert new technology) killed email".

One good source is Morgan Stewart's list of data sources on the state of email use and email marketing.

Another is EmailIsNotDead.com. I put up that site as a one-page factsheet with stats and article links you can show to anyone whose head is turned by sensationalist headline writers.

I deliberately stay out of "Is email dead?" arguments because I'm not convinced of the practical value.

But I understand that people with budgets to plan and/or fight for need hard evidence to support future investment in different channels. Hence the new site.

One point that does need to be made is that email is, obviously, not dead. But it is changing.

And will continue to change.

As will other media.

At the moment, for example, people aren't keen on getting overtly commercial messages through social media. I've been online long enough to remember people saying the same about banner ads on websites.

Buddhists will tell you that impermanence is a basic condition of existence. They're right.

[The series on the future of email deliverability continues next week.]

Future of deliverability: 2. The role of authentication

2009-10-21T09:29:00.002+02:00

[Be sure to read Part 1 of this series: User interaction]

OK, hands up if you don't understand how email authentication works.

(This is where I sincerely hope I'm not the only one holding my hand up).

The good news is you probably don't have to understand it. But your email marketing service or IT technicians should know how it works because your emails and/or email system need to support the authentication process. Here's why...

Authentication is not a deliverability solution, but...

A simplified definition of authentication is that it's the process by which the alleged identity of the sender of an email can be verified. It requires action at both the sending and receiving end of the email delivery process.

Senders must modify some technical domain records and (depending on the type of authentication) also modify the information sent along with emails so that receivers can run appropriate verification checks.

Interest in authentication comes from a need for more accountability in email. You can see immediately how verifying sender identities helps the battle against, for example, phishing.

If an email purports to come from PayPal.com, but cannot be authenticated, then organizations managing incoming email can tag it as suspicious or block it.

Only PayPal can properly authenticate their outgoing email as "from PayPal" because only they have access to the appropriate domain records etc.

Authentication lets us check if mail claiming to be from BigBrand.com really is from BigBrand.com.

But it says very little about the value of BigBrand.com's emails and whether those emails deserve delivery to the inbox.

So authentication can help stop people claiming your identity (phishing) but authenticating your email doesn't magically increase your delivery rates.

As George Bilbrey (President) and Tom Sather (Professional Services Director) of Return Path point out:

"Authentication is not about deliverability and that was never the intention. It's about phishing and spoofing. The intention was never to hurt (or help) deliverability."

So what's authentication doing in a deliverability post?

The use of authentication does have indirect impacts on your ability to get email delivered.

As Deirdre Baird, President & CEO of Pivotal Veracity puts it:

"It's very important to authenticate...not being authenticated doesn't necessarily hurt you, but being authenticated provides additional benefits that can help deliverability..."

So much so, in fact, that Chris Wheeler, Director of Deliverability at Bronto adds:

"In this day and age and with large-scale adoption at major ISPs, not having authentication is something you cannot get away with any longer."

There is already talk of authentication having a direct benefit to spam filter scores. And some email address providers are already flagging non-authenticated email with warnings.

Such warnings affect deliverability because trust is an important factor in determining how a subscriber interacts with your email. Remember, positive (clicks) and negative ("report as spam") interactions with your email play an increasing role in determining your future delivery success.

In Germany, Web.de and GMX webmail services now display "trusted" and brand icons next to authenticated email from selected senders. Gmail has begun something similar.

Suddenly, authentication starts becoming very interesting in terms of trust and brand building.

As it stands, though, the three main benefits of authentication for marketers are: brand protection, access to delivery resources and sender reputation.

1. Authentication helps protect your brand

Accountable email helps protect you and your customers from criminals attempting to send emails that pretend to be from your brand.

Bilbrey and Sather note that authentication as brand protection should be a top priority, and not just for big companies:

"If companies think they don't get phished or spoofed they are mistaken. Through the Return Path Reputation Data Network we get data on millions of messages coming from ISPs every day. And a shockingly high number of those messages are phishing and spoofing attacks."

"Companies are being spoofed and phished all the time. This is true even of relatively small, lesser-known brands. Authentication can limit the damage."

Wheeler adds:

"Brand affinity is put at risk when the potential fallout from spammers and phishers is substantial. Authentication helps avoid this."

2. Authentication enables access to information and services

Some ISPs require senders to authenticate email before they will provide a lot of courtesy delivery data. Jeremy Saibil, Director of Deliverability at Campaigner, says:

"...if you are not signing Yahoo-destined mail with DomainKeys (DK) or DomainKeys Identified Mail (DKIM) authentication then you are unable to participate in the Yahoo Feedback Loop. DK/DKIM signing is also a requirement for Gmail's new unsubscribe/complaint reporting function."

Feedback loop data lets you remove complainers from your list and provides critical information on how recipients are reacting to your different emails. According to Saibil, if you don't get access to this data, you "...are missing a HUGE piece of the puzzle required to judge the effectiveness of your campaigns."

And he concludes:

"I suspect every piece of informational candy ISPs will share with email marketers moving forward will have an authentication requirement."

3. Authentication supports the next phase of sender reputation

Perhaps the biggest delivery influence of authentication in the future will be in support for domain-based sender reputation.

In the ideal email world, those managing incoming email could build up a behavioral history for specific senders (e.g. BigBrand.com), and judge their emails accordingly. You would have a reputation associated with your domain name.

This is good for receivers and good for you, since your reputation is more directly linked to your actions and its independent of the location (IP address) from which your emails are sent.

[Hint: Part 3 will look at the dawn of domain-based reputation in more detail.]

Clearly, this kind of reputation system depends on authentication. Receivers can only use domain-based reputation if they can be sure of the provenance of the email. As Baird puts it:

"Authentication is a pre-requisite for taking advantage of domain-level reputation at major ISPs."

So what authentication standard should you implement?

Of course, there is more than one way to authenticate your email. Historically, the main standards are:

Sender Policy Framework (SPF) and Sender ID
DomainKeys and DomainKeys Identified Mail (DKIM)

Which do you go for?

SPF/Sender ID are relatively easy to implement and important for Microsoft webmail properties. But DKIM is emerging as the real must-have standard. Saibil says:

"DKIM in particular is the authentication standard that is really going to matter moving forward..."

To learn more about DKIM, see these articles:

Searching for Truth in DKIM by J.D.Falk
DKIM: Not Shiny, But Very Important, also by J.D.Falk
How hard is it to deploy DKIM? by John Levine
What do you want out of DKIM? by Al Iverson
DomainKeys Identified Mail (DKIM) Grows Significantly by Jim Fenton
DKIM: what it's not by Laura Atkins
Determining the DKIM Value Proposition by Dennis Dayman
Franck Martin on the Future of DKIM and Domain-Based Reputation, also posted by Dennis Dayman

Should you panic if you're not authenticated?

Not for the moment. Bilbrey and Sather tell us:

"As it stands today, authentication is at barely 50%. If ISPs started blocking all unauthenticated mail the false positive rate would sky rocket."

But all our experts advise ensuring authentication ASAP. As Baird states (with my emphasis):

"As ISPs continue to evolve their filtering mechanisms, authentication will play an even more important role in determining the good actors from the bad actors. By setting up email authentication on your mailing domains and IPs you are building a reputation history for the future."

Part 3: domain-based reputation

More on deliverability

Future of deliverability: 1. The role of user interaction

2009-10-16T09:45:00.002+02:00

The growing number of specialist deliverability services and consultants is no coincidence. Email deliverability is a tough topic to stay on top of.

Those managing incoming email (ISPs, webmail services, software manufacturers etc.) are constantly modifying how they sort those messages so that bad email is kept out of each user's inbox.

That's a good thing. But what makes a "good" email? Just where is deliverability heading? What do marketers need to plan for?

Four deliverability experts kindly shared their understanding with me for a series of posts covering the biggest trends in deliverability and what they mean for marketers.

This first part looks at the role of user interaction (engagement).

[If you're not familiar with the basic terminology in deliverability, don't worry: here's a simple glossary.]

User interaction and reputation

We know that ISPs and others are using sender reputation to help determine whether an email deserves delivery to, for example, the inbox or the "junk" folder.

Various factors get taken into account when building a reputation for a particular sender (usually defined as a source IP address) and one focus is the number of spam complaints received.

A spam complaint is the classic example of an engagement metric or user interaction contributing to sender reputation.

By clicking on a "report this as spam" button, the user is making a statement about the perceived value of the sender's emails. These statements then flow into the calculation of reputation.

But spam reports are not the only example.

Increasingly, the likes of Yahoo! Mail and others are looking at all sorts of positive and negative user interactions to build up their overall picture of your sender reputation.

Examples?

George Bilbrey (President) and Tom Sather (Professional Services Director) of Return Path offer three examples that one or more top ISPs are already using:

Do users open messages from the IP/domain?
Do users click on links coming from the IP/domain?
Do users mark wanted email that appears in the spam folder as "not spam"?

They add:

"Another technique to measure engagement is to create a large panel of trusted users...The users are given a sample of their messages and asked to vote each as spam or not spam. This technique is most notably used by Microsoft."

Bilbrey and Sather note that the "this is not spam" metric has been in use for over three years:

"It's the primary way that many ISPs get a feel for whether they are making a mistake in placing a message in a junk/bulk folder. Most of the large ISPs that have built their own reputation systems are looking at this."

Will this approach see wider application?

So, is this engagement approach to email filtering likely to spread?

Yes, says Jeremy Saibil, Director of Deliverability at Campaigner. He continues:

"In fact, it already has been happening on a broad basis...ISPs have and will always leverage this type of information to help serve their clients better.

Deirdre Baird, President & CEO of Pivotal Veracity adds:

"...top ISPs that often account for 50% to 75% of B2C marketers' lists such as Yahoo, AOL and Hotmail are measuring engagement (clicks, forwards) and disengagement (persistent ignoring) in overall and individual-level folder placement decisions." (my emphasis)

But what about smaller or regional ISPs without the resources of a Microsoft? Are they applying engagement metrics, too?

Chris Wheeler, Director of Deliverability at Bronto says once the big players prove the usefulness of the approach, others will follow:

"Just as in the past with Feedback Loops and authentication, a wait and see approach by the smaller ISPs is usually the way programs become more widespread."

Bilbrey and Sather agree:

"Smaller ISPs don't really have the power to do it right now. But even some second-tier ISPs are starting to experiment with less sophisticated versions. So we expect to see a lot more of this in the coming years."

Should you worry?

The growing importance of engagement metrics in determining your delivery success makes you even more accountable to your subscribers. Wheeler tells us:

"The interesting thing here is that senders, once again like with the TiS (This is Spam) button, will be at the mercy of their recipients. If marketers send email that is received well and opened and/or clicked on, their good deliverability will reflect this."

He continues:

"It really puts the onus on the creative and marketing strategy to capture recipient engagement. If you send email out that doesn't generate complaints, that has been sufficient up to this point. Moving forward, you'll have to actually drive your recipients to action as well."

But surely any half-decent email list should easily meet the engagement criteria that might be used by ISPs? Not so, say our experts.

Baird warns:

"Although marketers may make all the necessary technical configurations to their outbound mail stream, mailing practices such as over-mailing may cause list fatigue that will lead to an erosion of engagement and ultimately cause messages to be placed in the spam folder..."

And Bilbrey and Sather note:

"Too many marketers are still satisfied with relatively low response rates. If they have a big enough list they can make the math work even at very low click rates. But that is now starting to have a deliverability penalty."

...and they cite the example of a top online insurance company:

"They aren't scraping email addresses off bathroom stalls. The frequency is fine. They are even doing lifecycle segmentation to get the right messages to the right recipients. All way more than many marketers do, frankly. But they are struggling to get to the inbox at AOL and Yahoo! because of their engagement metrics." (my emphasis)

Ultimately, it's down to the quality of your email marketing and actively managing those engagement metrics. Saibil says:

"I'd expect that anyone running a decent email marketing program would be very well versed in these metrics. Most likely they are already proactively adjusting their programs should positive interaction metrics drop off."

And he has a tip for us:

"...one of the most important deliverability tips I preach is ensuring you have consistent visibility of your own campaign stats, ensuring you have a baseline for measuring both positive and negative engagement metrics."

It's not all or nothing

Baird also warns that the use of engagement in sender reputation gives ISPs the information they need to direct emails on a recipient-by-recipient basis:

"In the past, if you had a good reputation based on low spam complaints, good list hygiene, and a few other metrics, you could count on inbox delivery. Now, you can have all those things in place, and may have a high delivery rate, but a portion of your list - the disengaged customers - will not be receiving your mail in their inboxes like everyone else is."

"The most important thing to keep in mind here is that even if a marketer has a great email program, stellar reputation and "will pass", the individual's explicit and implicit preferences will still take precedence over which folder your mail is routed to." (my emphasis)

There's a positive flip side to that of course, as Baird explains:

"...if a lot of your mail is being sent to the spam folder, an individual's positive engagement with your messages will override your reputation elsewhere and ensure your messages are routed to that individual's inbox."

The key takeaway is this:

Acceptable response rates in the past (from a profit perspective) may not be sufficient in the future: engagement becomes a goal in its own right, thanks to its impact on customer behavior and deliverability.

Part 2: The role of authentication

Still to come: domain-based reputation, reputation and B2B lists, certification, and great links for further reading. Stay tuned...

More on deliverability

Video email: new technique + take a poll

2009-10-14T21:27:00.001+02:00

Looks like our video email toolbox might have another addition.

Dave Greiner of Campaign Monitor has successfully tested using the <video> element available within the proposed HTML5 specification.

That sentence already exceeds my technical knowledge, but basically the code he used either let a video file play directly in the email (e.g. in Apple Mail) or displayed a clickable image instead (e.g. in Outlook 2007). See the details here.

Pretty impressive, notwithstanding concerns as to how email clients and webmail interfaces will handle HTML5 elements in the future.

We now seemingly have five video email options:

1. A still or other image linking to a playable video on a website
2. Video .gifs (an advanced application of animated .gif technology)
3. CertifiedVideo (embedded video with limited ISP reach)
4. Embedded YouTube videos (only works in Gmail under specific circumstances)
5. The HTML5 <video> element

[There are also video email services out there that address the technical challenges of the medium, for example by optimizing streaming processes or providing viewing analytics.]

Now we know we can do video email, we can start asking "which technique should we use?" and "what tactical and practical issues do we need to worry about?" and (most importantly) "should we use video email at all?". Previous posts dealing with these topics are:

Video email: current practices
Video email: 8 recommended practices
Video email: when to use it

Finally, I'd be interested in whether you've put video content into your emails.

Here's a one-question poll on that (feed readers may need to click here to see it). Feel free to use the comments to expand on your experiences. This is new stuff for most of us and "best" practices are still to be established...

Thanks!

Before you start a reactivation campaign...

2009-10-09T16:03:00.001+02:00

A common theme in recent posts has been the need to get people interacting with your emails. Um...not a startling insight, really.

But it's not (just) because a view, impression, click is the start of the path to some kind of conversion. It's also because ISPs are taking this interaction into account when deciding whether your emails should go to the inbox or the spam folder.

As interaction becomes important for its own sake, so the risk associated with "inactive" subscribers grows. A big heap of unresponsive subscribers tells ISPs that "these emails aren't worth putting in the inbox".

Which explains the renewed interest in "reactivation" campaigns designed to get dormant recipients opening and clicking again.

As so often happens, our approach to reactivation is compromised by the seductive appeal of semantic simplification. (I have always wanted to write a sentence like that).

Translated into English, it means we think of inactives as a uniform group of people who ignore our emails, who are blind to our messages.

We just need to give them a gentle tap on the shoulder and...whahay!...they'll rediscover the email gems we send and turn into born-again subscribers.

If only it were that simple.

At this page, you'll find links to 27 articles from email experts with advice on reactivation campaigns. But before diving into those, here are three critical questions to ask yourself...

1. What kinds of inactive subscribers do you have?

This is the crunch question. How do you define "inactive", bearing in mind that not all inactives are created equal? Or are even inactive!

One common split is between those who have never opened or clicked on an email and those who responded in the past, but haven't done so for whichever period of time you determine labels them as inactive.

And you can split these down further. Here are three plausible subgroups for that first segment:

1. Have never seen your email, as it doesn't get past the spam filters.
2. Only signed up to access a sweepstakes or one-off incentive and have no interest in your products/services.
3. Signed up expecting something else.

The right reactivation strategy depends on your ability to identify the true causes of inactivity.

Subgroup 1 is a deliverability problem. You need to solve that first. Then you're in a mailing to an old list scenario, rather than a reactivation one.

Subgroup 2 probably isn't worth trying to reactivate.

And Subgroup 3 suggests you need to modify your sign-up copy and welcome messaging as a priority before thinking about reactivation.

Our second segment (those who have responded in the past) also likely has various subgroups.

Maybe some folk simply changed interests, which implies you need to find the right content/offer to match these changes.

Others maybe tune you out because you send too much email. Then a simple frequency adjustment is required.

2. Are your inactive subscribers really inactive?

Now let's make it even more complicated. Some of those inactives aren't inactive at all. Here three examples:

1. They may be getting your emails but responding in some way you're not measuring, perhaps through another channel.

Here a quote from Jake Holman:

"When we compared the data, we found the customer was receiving an email...picking up the phone almost immediately and ordering huge amounts of products."

And another from David Baker:

"Are they really dormant customers or is there a channel shift in place? Are they converting through search, media and you aren't getting credit as the 'last click'?"

2. They may be interacting with your emails.

Someone may assiduously read every word you write, but never register an open since they use a mobile device or have their email client set up to read the text version of your multipart message. Without images displaying, no open can register.

This scenario is particularly likely if you send the kind of emails that don't necessarily demand a click to access the value (such as newsletters featuring full articles).

3. They may appreciate your emails but are just waiting for the right time to respond.

Dela Quist has long argued that a significant number of inactives want to be on your list and are just "unemotionally subscribed" as opposed to the commonly assumed "emotionally unsubscribed":

"...they do want to receive your emails, but don't need your content or offer yet. They would prefer to ignore your messages until they are ready to buy, because it is easier than unsubscribing and having to remember your url or Google you at a later date."

When MarketingSherpa actually called up non-responders to find out why they weren't responding to emails, the most popular answer was "I like your email. Don't stop sending it. I may not always have time to read it, but I want it".

As Anne Holland explains in the same article:

"They may not click on your email, or even open it. But they see your brand name in their in-box, as well as a subject line. Those two items alone may trigger a delayed response later."

And this is even before we get into any arguments about how long a subscriber needs to go without clicking or opening before they can be considered "inactive".

John Caldwell, for example, reminds us that your business model drives this definition.

You'll often see six months of inactivity cited as a good definition of an inactive subscriber, but as John notes:

"I can go 9 months without ordering flowers, but will always be back for the same special occasions. How are you going to win me back if I never left?"

The real key is to find ways to identify the different kinds of "inactive" email addresses so you can adjust the reactivation approach accordingly.

All of which is easier said than done. Those 27 articles will help.

Even if you use just one reactivation tactic (such as a "please renew your opt-in" campaign), at least adjust your approach and copy to account for the possibility that some subscribers may be surprised to find you think of them as inactive.

3. Do you have a plan for afterwards?

When you get a response out of a previously dormant address, what will you do to keep them active?

If you put them back in your active file, aren't you going to send them the kind of emails that drove them to go inactive in the first place?

If that scenario applies, then the reactivation campaign is treating the symptoms but not the problem. Solve the problem, then try and regain lost attention...

Again, this goes back to understanding why people go inactive. If your messages were no longer relevant, you need to address that issue before you try and get them back into the fold.

Which is why an ideal time for a reactivation campaign is just after you've upgraded your program to deliver more value to recipients.

The dangers of subscriber loyalty. Eh?

2009-10-05T11:42:00.001+02:00

Having loyal subscribers makes things easier. Once you've earned that loyalty, the pressure comes off the email production process.

It doesn't mean you can sacrifice on the value you send in your content and offers. That way lies ruin. But it does give you a little leeway in terms of packaging.

What starts to happen is that such things as subject lines and sending times diminish in importance. Provided your email is easily recognized, people will look for it, find it and consume it.

Loyalty drives attention

If, for example, you send out a B2B newsletter on a big public holiday, you'd expect responses to that newsletter to fall below average. People aren't at work and we know most opens are normally recorded in the first few hours after a send.

Here the numbers from a real example sent out on September 7th, a big US holiday (Labor Day):

The results are not that different to a "normal" send day. People simply open the email the next day:

How can this be a bad thing? Well, it isn't. If you're lucky enough to have a loyal audience, this is great news. But there are dangers associated with such loyalty.

Danger 1: Complacency

Loyalty, like permission, is not permanent. It depends on continuing to deliver the value that turned them loyal in the first place.

Loyal subscribers may be more forgiving of the occasional dip in quality. But they have their limits, too. You don't want to risk falling below the extinction threshold.

Danger 2: Not everyone's loyal

As a list ages, new subscribers make up an ever-smaller proportion of the total list size. So their response to email elements like subject lines, offers and send times is effectively masked if you have a particularly loyal subscriber base.

Your current email approach may be turning new subscribers into ex-subscribers or inactive subscribers without you spotting any big shifts in response rates.

Danger 3: Clique speak

A related problem is to assume subscribers are all very familiar with you and your emails. This can lead to clique speak, using jargon or an insider style that long-term subscribers can relate to but leaves newcomers confused or excluded.

Another problem is that you assume everyone read the last 50 emails. But for some people, this email is their first from you. Oblique references to past content simply leave them confused.

These dangers mirror problems that online forums encounter when forum growth plateaus. Newcomers can feel like they're intruding on a private gathering where everyone knows everyone else.

While some newcomers are incented to gain access to that insider clique, others simply won't bother to familiarize themselves with the tribal rules and wander off to somewhere more welcoming.

See if you have an issue

If you do have a loyal audience, pull out statistics for new sign-ups and see how they compare to the average: look at response and unsubscribe rates.

Under normal circumstances, these should be better than the norm, since the original interaction that led to a sign-up (like a purchase or website visit) is still top of mind. New subscribers should be more engaged than average (to begin with).

If they're not, perhaps you have a loyalty "problem".

Solution 1: Consider welcome streams

You probably have a welcome message for your email program. Consider broadening this into a series of welcome messages to ease new subscribers into familiarity with your messages before adding them to your standard email flow.

These links will help with the theory:

Tactics 3 and 4 in this post outline why you do this
Best practices for initial emails after sign-up
The power of welcome emails

Some examples:

An award-winning welcome series by REI and Smith-Harmon
One bank's approach to "onboarding" emails
An evaluation of how Netflix does it

For other tactics designed to prevent newcomers abandoning your list immediately, see the article Avoid subscriber's remorse.

Solution 2: Get a fresh perspective

Consider showing your messages to someone who is not familiar with them. They can help pinpoint elements that make no sense to the newcomer.

You don't want to remove every "insiders only" element from your emails. After all, it's often those elements that help your emails stand out and cement loyalty in the first place.

But a fresh perspective may identify small changes in tone, style, language, vocabulary etc. that retain the character of your emails but make life easier for those new to your list.

Solution 3: Continue to work on improvements

If your tests are not throwing out clear winners, look at the results of each test for different subscriber groups. While subject line changes may not move the needle much for long-term active subscribers, they may be showing you how to get more attention from newcomers.

How to get the click (even when you can't)

2009-09-29T14:09:00.000+02:00

Most people don't click on emails. And most marketers don't care.

You can pull out all the tools in the email marketing shed: copywriting, subject line optimization, calls to action, trigger mails, segmentation...but you still can't avoid the problem that most people don't click.

Why? Because right now I'm just not interested in buying a new pair of trainers or reading about changes to tax laws.

But who cares, right?

The click is the first real action on the road to a conversion. Whether that conversion is a sale, download, registration, donation, whatever.

If somebody isn't going to convert, then why would you worry about getting the click?

Here are four reasons...

1. Email isn't only about conversions. It's about getting people to interact positively with your brand or organization. The deeper this interaction, the greater the impacts on the "relationship" between sender and recipient...the more attention and influence you gain.

2. We are creatures of habit: if we found something clickworthy in one email, we are more likely to pay attention to the next one.

[And vice versa, as explained in the recent post on the slow death of your email...]

3. A click takes people to a website, which is a much richer display environment. It gives you more opportunities to engage the reader and drive some kind of positive (inter)action.

4. Positive interaction with your emails sets you apart from "bad" emailers. According to a recent Pivotal Veracity report, webmail services are increasingly taking user-email interaction (including clicks) into account when determining whether a sender's emails are worthy of delivery.

So clicks are worth having in their own right, but here's a crass example of the problem. These are the last few emails received from a domain name registrar:

If I'm not interested in actually buying a domain name, I'm not going to click on these emails.

[Incidentally, even if I am, there's no sense of urgency, since clearly I can wait a few days for the next discount to come trundling along.]

So we want a click, but we come back to our original problem that what we feature is simply not relevant to the majority. So what can you do? Here some ideas...

Offbeat items and humor

Newsletter expert Michael Katz writes:

"...I find this consistently across all types of newsletter, my own and those of my clients. No matter how "serious" the subject matter, the highest clicks are always the diversions."

I can echo that for my own newsletter. Links to such things as Darth Vader's inbox or a talking email regularly score more clicks than even the most insightful marketing articles.

Of course, humor is tricky. We can't all be Ricky Gervais. But you can think of these "diversions" in other ways. Amy Hamilton, for example, notes:

"I find that the emails I tend to click on and forward the most are the ones containing the most outrageous products, just for the humor of it."

...and features some examples from Buckle, Forever 21 and Abercrombie & Fitch.

Ancillary calls to action

Your email usually focuses on getting people to do something very specific (buy the product on offer, read the article, register for the webinar).

Provided they don't distract from your main content, additional calls to action can pick up clicks where the main focus is not relevant to the recipient.

Navigation bars, for example, are menus of links to one side of the main content:

In his report on the topic, Chad White says navigation bars:

"...give your subscribers a clear and familiar path to engage with you even when they're not interested in your email's main message."

Another example is outlined in this post: adding footer links to content from previous emails lifted the number of clicks per opened email from 0.47 to 0.61.

Video animations and "click to find out..."

I've talked a lot recently about video email, and one post cites various examples where video lifts clickthroughs quite significantly.

Is there a broader principle at work?

Anne Holland reported on a website video test and noted:

"...perhaps being forced to interact with the video to hear sound got more prospects into 'interaction mode' so they were then more willing to click a 'join' button next."

Perhaps, then, part of the click appeal of the audio-free video .gif approach in email is a simple desire to hear the sound. The click puts the reader in interaction mode and leads them further down the conversion road.

The only Ology I ever studied was biology, but it seems reasonable to suggest that "teaser" video might work just as well to gain clicks as the teaser email summaries used to get people to clickthrough to online articles.

You might make broader use of the teaser concept, for example by featuring mystery offers ("click to reveal the discount offer") or linking to the coupon code ("click here to get your personal coupon code").

Teaser summaries for articles enjoy widespread acceptance among recipients, because they understand you can't fill an email with too much article text and they enjoy the ability to quickly scan and choose what to read.

The same might not apply to those kinds of teasers where there is no clear reason why you can't just put the content in the email.

So those kinds of teasers need to be used sparingly, otherwise you give the impression that you don't value the time and attention of your subscribers.

The content approach

I've asked before whether email marketers should become content publishers to keep people engaged and solve the "most don't click" problem.

There are different ways to approach this. For example, you can:

...go the newsletter route and focus mostly on engaging content, like Backcountry does with its newsletter.
...tag content onto promotional messages, like Columbia Sportswear does with the video review feature in this email discussed by Dylan Boyd.
...integrate promotional messages into content, as in this simple self-tanning how-to from Bliss.
...supplement promotional mails with tidbits of free information, as Linda Bustos explains here.

Feedback, surveys, competitions, reviews, polls, loss leader offers

Of course, building interaction into email dates back to the first use of a question mark in the text. So let's not overlook tried and tested ways to get people to respond without having to follow a "shop now" or "continue reading" or "register today" link.

Some examples:

Jake Holman describes the great copy used by MarketingProfs to solicit feedback on an online seminar
Blue Tent Marketing won a MarketingSherpa award for their holiday-themed customer survey email
...as did Air Canada for their Dream Destination email competition, which carries the tagline "Dream. Click. Win a trip." (my emphasis!)
SmartPak send out automatic emails 14 days after a purchase soliciting product reviews. The tactic increased reviews sixfold.
Poll scripts don't work in emails, but you can ask a single question and provide a selection of answers in the form of links. People vote by clicking on a link and your tracking data records the winning answer (and who voted for what). David Dennis explains the technique and how you might use it to build out your subscriber profiles.

Note that all these efforts drive engagement, but serve other goals too: data collection, profiling for later segmentation, content generation, etc..

Finally, an earlier post raised the possibility of loss leader emails: ones that feature extravagant offers or incentives that attract plenty of clicks to gain engagement wins at (perhaps) the expense of short-term profits. Amy Hamilton has some examples.

Any other suggestions for encouraging clicks and interaction?

Famous inboxes #3: Sauron

2009-09-24T12:25:00.000+02:00

For those new to the blog, every couple of months I put email marketing to one side and do something different. Like today. Non-geeks move quickly on.

After a sneak look at the email accounts of Darth Vader and Lord Voldemort, we now get a chance to see what Sauron gets up to when he's not chasing hobbits in Lord of the Rings.

New source of email marketing news

2009-09-23T14:23:00.001+02:00

I've pointed you to OnlineMarketing.info before, as it's where I put up a custom search engine that limits Google searches to 360+ hand-picked email marketing resources.

The site now also features news and article headlines on the start page.

These headlines are updated every 1-2 days and link to those new posts and articles from around the Web that I believe are of particular value or importance to email marketers.

I regularly review some 140+ blog and media site feeds and this is where the headlines are drawn from.

The sites covered by the search database are also now listed for all to see.

Thanks go to readers "That Guy Steve" for the "latest headlines" suggestion and David McLaughlin for the "site list" suggestion. Sorry it took me so long to implement them.

As always, any suggestions for sites to add to the search database or new blogs to follow are very welcome.

Subject lines: Amazon's lessons on discounts and frontloading

2009-09-18T10:43:00.000+02:00

Last month I ran three posts on what 187 emails from Amazon can teach us about subject line length, branding and personalization.

In a feedback poll, 84% of you asked for one more on the topic, so here it is...

Use of discounts and prices

As a retailer, Amazon inevitably uses email to offer discounts on individual products or product groups. These discounts vary from 10% to 90%.

Since the pulling power of a discount increases with the size, bigger discounts get more frequent use in their subject lines:

But what about those few times when a 10% discount features in the subject line?

Isn't 10% too weak for the modern bargain-filled inbox, especially when your other emails are touting 40%, 50%, 70% discounts?

This again illustrates the idea that a subject line is the sum of its parts and context dependent.

The less targeted the email, the more Amazon has to compensate with a bigger discount to encourage attention. So you'll find the big discounts featured in subject lines with broader product group promotions:

...and the smaller discounts where a targeted recommendation (based on past purchases) means the discount doesn't have to do the heavy lifting. It's icing on the cake:

Heavy discounts on a highly targeted email presumably make little sense for Amazon. If you're recommending a very targeted up- or cross-sell, people will likely buy anyway or only need a small discount to be persuaded. Larger discounts carry a big revenue penalty and the additional lift in response doesn't compensate.

Where targeting is weaker, however, Amazon uses the bigger discount to provoke interest and encourage people to explore the individual offers or product categories featured in the email's body.

Since we're often told that subject lines need to be specific and reflect the email's content, does putting actual offer prices in the subject line make sense?

Amazon doesn't seem to think so. Only a handful of those 180+ subject lines feature actual prices. This likely reflects the idea that the subject line is not driving action directly, but starts a sequence of events that will eventually lead to that action.

In other words, the subject line should not get the recipient to decide immediately whether or not to click, buy or make some other desired response.

Instead, it should persuade them to investigate further so you can exploit the richer environment of the email body and landing page to get the desired response.

[Another issue is the role that currency symbols and prices in subject lines might play in content filters.]

If you accept this approach as valid (some retailers do feature prices in subject lines more often), there are still two exceptions used by Amazon.

The first is to use price as an indicator of attractive prices in general for a particular product group. As Amazon does here:

The second is where the actual price really is a deal closer.

Frontloading

We've covered some of the main "hotwords" that Amazon uses to grab attention: brand, personalization, discount etc. But what do they put first?

The further back a word appears in the subject line, the less likely it is to get read. If for no other reason, then because email clients and webmail interfaces limit the number of subject line characters on display.

So marketers tend to put their "hotwords" as near the front as they can. Here are the first and second most popular subject line beginnings at the three different Amazons I looked at:

Amazon.com
1. Brand name, e.g. "Amazon.com: Bestsellers in Biographies and..."
2. Discount, e.g. "Save 32% at Amazon.com on..."

Amazon.co.uk
1. Discount, e.g. "Huge savings on...", "Up to 70% off...", "Save 26% on..."
2. Brand name, e.g. "Amazon.co.uk: Find great offers"

[Incidentally, since the analysis was done, I'm getting more emails from Amazon.co.uk that lead with my name: "Mark Brownlow: Save up to 70% on..."]

Amazon.de
1. "Jetzt neu: ..." (translation: New release or New in stock)
2. Various...

While some differences likely stem from my different purchase patterns at each Amazon, it's a reminder that different audiences have different hotspots.

Perhaps the Amazon.com brand has a particularly powerful pull in the USA, UK shoppers are after bargains and Germans like to have the newest thing?

The point is simply that there is no universal "winning" subject line approach. The words you use should adapt to different audiences. Just as you might write different ad or landing page copy depending on who you're targeting.

When experts talk of email segmentation they usually refer to sending different content/offers to different groups of subscribers. But you might apply the concept equally to the presentation of that content or offer.

So the same offer or content goes out to the list, but different segments get different subject lines. You might lead with the discount for the "bargain hunter" segment, the newness for the "want to be first" segment or your brand for those who see you as their trusted adviser.

OK that brings to an end my interpretation of what the marketing brains at Amazon are trying to do with their subject lines...do you agree? Why do you think they do what they do?

More on subject lines | Tags: email marketing, subject lines

Top email marketing info sources for 2009

2009-09-16T19:20:00.002+02:00

Time for a 2009 update to my annual guide to the top people and places to go to for email marketing information.

There are now hundreds of great sites, blogs, newsletters and writers in the email marketing world. So these are just my personal "favorites"...confined to info available in English and admittedly US-centric.

Please do list your own favorites in the comments.

So, in no particular order...

Media sites

ClickZ was one of the first marketing media websites to publish an email marketing column. They now have five, featuring many long-time industry stalwarts.

All are good value, but my particular favorites are the optimization (Derek Harding, Ed Heinrich) and delivery (Stefan Pollard) columns.

MediaPost's Email Insider hosts various excellent columns, written by experts with strong backgrounds in the subject area.

If I was forced to name favorites, I'd go for Loren McDonald's column, which probably comes closest to the way I think about and use marketing email, and the design inspiration provided by Lisa Harmon and colleagues.

Check the iMedia Connection site for regular articles from another slew of email experts like Wendy Roth, Spencer Kollas and others.

MarketingProfs offers many great resources, but the email newsletter "Get to the Point: email marketing" is perhaps best. You get summaries of email marketing articles at the main site, plus highlights of useful posts from other sources.

For a more UK-oriented view, check out Econsultancy's email marketing articles.

And, finally, I know marketers who read Ken Magill's columns just to make sure they're not in them. He is just about the only one out there willing to burst the cozy mutual backslapping bubble that all industries wrap around themselves.

Stats and reference works

MarketingSherpa's Email Marketing Benchmark Guide has achieved iconic status in the industry. The main Sherpa site is also a good source of case studies and other reports.

The EmailStatCenter.com website is the place to start looking for public stats. It's the brainchild of Simms Jenkins, who passes on his broader email marketing experience through an iMedia Connection column and The Truth About Email Marketing book.

Other recent books on email marketing include Successful email marketing strategies and Email marketing: an hour a day.

ExactTarget's Morgan Stewart is one of the few experts out there with an understanding of email, wider marketing channels and statistical analysis. I value his blog posts very highly.

Twitter accounts

I have dozens of email marketing Twitter accounts listed here. And many of the highlighted names in this post are, of course, worth following at Twitter, too.

A special shout out, though, to Fred Tabsharani, Captain Inbox and John Caldwell who do a great job of passing on useful links to their followers.

Blogs and newsletters

You can get an OPML file of all the 140+ blogs I subscribe to here, but some highlights:

Chad White operates the Retail Email blog which is a super public source of insight into what the big retailers are doing with their emails. Also see his column at MediaPost.
The MailChimp blog often comes up with great insights and stats, such as numbers on the effectiveness of different subject lines and advice on what kinds of address lists can be used for email marketing.
eROI has several blogs which are always great and regular sources of practical tidbits, event reports and musings from the front line of email marketing. My faves: Return on Subscriber by Alex Williams and The Email Wars by Dylan Boyd.
The Bronto bloggers have impressed of late with their concerted effort to bring useful hype-free info to the community.
The Blue Sky Factory blog stands out because the people there are also very clued in to social media and the implications for email. And because it features the irreverent DJ Waldow.
Finally, try the Alchemy Worx newsletter. It takes common topics and offers a new perspective or advice that challenges what we all thought was true. CEO Dela Quist is also the only reader who doesn't leave comments on my blog, but actually rings me up to debate a point.

Small business

Most of the resources mentioned in this post are for people dealing regularly with email marketing. Here I'd like to point out some resources that address small business and those dealing intermittently with email marketing...

If you write an email newsletter and haven't read the articles by Michael Katz, you should.
Janine Popick's blog posts at Vertical Response take a small business focus and go beyond the mechanics of email marketing to cover other business aspects relevant to the "small" emailer.
The folk at AWeber are very good at highlighting key stuff for small business and those with smaller lists and autoresponders. Look out particularly for Justin Premick's posts.
The StreamSend blog is a constant source of quick tips for email marketers.
Constant Contact has a huge small business user base: see their Learning Center for blog posts, articles, webinars etc.
Campaigner's blog is a regular series of quick tips on email marketing and small business (in particular).
Tom O'Leary writes the Messaging Times blog for Infacta.

Asking a question

The Email Roundtable is a marketer-dominated discussion group with some of the brightest folk in email marketing on board. The discussions deal with the day-to-day strategic and tactical issues faced by email marketers and are characterized by their pragmatism, experienced insight and friendliness.

The Email Marketer's Club is probably the biggest networking site out there for email marketers and includes a forum and many other useful resources. It was founded by Tamara Gielen, who also runs the BeRelevant blog which regularly aggregates links to the more useful email marketing content out there.

Bonus: Deliverability

Return Path are never shy about commenting on issues that coincide with their interests. But they offer a different vendor perspective on their blog as they're not a traditional email service provider. In particular, Stephanie Miller gets through my trust barrier and you'll find her articles at places like ClickZ and MarketingProfs.

As well as the occasional appearance on his employer's blog, Al Iverson pops up in various places to display his in-depth knowledge of spam, deliverability and blacklisting issues.

Laura Atkins is an independent consultant who isn't carrying around any particular vendor bias when it comes to delivery issues. Get her advice and links from her blog.

Led by Dennis Dayman, the Deliverability.com blog pulls in contributions from a veritable who's who of the email (delivery) world. A special mention for two of those names...

Matt Vernhout also runs the Email Karma blog with deliverability news.
I don't always agree with Andrew Kordek but he's an important voice because he provides the crucial perspective of the frontline marketer with real-world pressures. A pragmatic counterweight to "best practice" advice from those who don't send email themselves (many vendors) or who operate in a non-retail environment (like me).

Traditionally, Pivotal Veracity have kept themselves to themselves, but you'll find Deirdre Baird and colleagues quoted in media articles: they have a very profound understanding of deliverability issues, so pay attention to them.

Bonus: Design

There are so many great HTML email design resources (e.g. Email Standards Project, Style Campaign, Smith-Harmon blog) that there is another post dedicated to that topic alone.

OK, over to you to list your favorites.

More email marketing resources | Tags: email marketing, online marketing

The slow death of your email (and how to stop it)

2009-09-11T09:23:00.000+02:00

We'd all like to be running one of those email marketing programs that deliver the Hollywood results you read about in case studies.

Most of us, though, are trundling along with unspectacular programs delivering solid results. And have been for a while. Emails work, even if you're not a database marketing hero.

But your program might be dying a slow death, thanks to the way subscribers behave and the new email landscape we now live in.

Here's how open rates have changed over the last 2 years on a small B2B newsletter I run:

It doesn't take a statistician to reveal that open rates are in decline.

Some of that is down to the growth of image blocking (though that excuse is getting a little old). Some of it might be delivery issues. Some might be a decline in the quality of the emails (hope not).

Slow death factors

But there are two factors ensuring that declining responses will likely affect every email program that keeps doing more or less the same thing as before. Like mine.

First, there is ever-increasing competition for subscriber attention. Both from other emails and all the other ways people now have to get information and promotions (most notably social media).

Second, the longer someone is on your list, the less engaged they are (on average). Here's the chart to prove it, which looks at open rate by year of sign-up for the most recent newsletter:

This slow decline in response rates due to such factors as aging and competition doesn't get much attention, because list growth usually compensates.

So while the percent of people responding drops, the absolute number of responders stays steady or even rises. Example:

The August 27th, 2007 issue of the above newsletter got an open rate of 47% and a click-to-open rate of 30%, which meant 160 clicks.

Two years later and the open rate drops by over a third to 30% for the August 24th, 2009 newsletter. Click-to-open rate is still 30%, but the list is now three times the size it was.

List growth more than compensates for the drop in open rates: the number of clicks is 332.

Actual responses have doubled, even though relative attention has fallen.

So why worry? Surely it's the end result that matters?

Think beyond list growth

There are two big issues on the horizon here.

First, using list growth to compensate for declining response rates is not a long-term strategy.

The older and bigger your list, the more new addresses you have to find to compensate for list aging. And the low-hanging fruit is likely already gone: your frequent visitors or customers are already on the list.

You have to work harder and harder on list acquisition, investing more and more resources. At some point, it won't be enough.

You may also end up seduced into questionable list acquisition practices that see subscriber quality drop and lead to permission and spam problems.

Engagement is also a deliverability issue

The second issue is deliverability.

A while back, people in the industry were predicting that those organizations managing incoming email (particularly the big ISPs) would broaden the list of criteria used to define spam (unwanted) email to include how people interact with a sender's messages.

If recipients aren't interacting positively with an email, this indicates he message is unwanted and the reputation of that email's sender suffers.

This is no longer speculation or theory.

Consider this quote from a recent report on a meeting with Yahoo's anti-spam czar:

"Yahoo stressed that the key to inbox delivery was to work towards positive engagement."

Among the signs of "positive engagement" cited: "Members are opening messages...members are clicking on the links in the message."

In other words, a decline in such numbers as open and click rates not only hurts responses, but might now lead to a poor sender reputation, in turn leading to deliverability problems.

Compounding the issue is the prospect that future inboxes will also include more tools allowing users to quickly identify or filter out those messages that are not spam but are also not that important (i.e. not open or clickworthy).

Ouch.

[You'll find more commentary on these sea changes in the delivery landscape from, for example, Stefan Pollard, Jeff Rohrs and George Bilbery.]

The obvious solutions aren't enough

So what can you do?

If you keep doing the same thing, your response rates will fall, you'll be under a lot of pressure to grow your list and you may run into new deliverability problems.

The obvious and glib answer is to certify your emails and get into all those advanced tactics that we keep putting off because things are still pretty rosy. Now is the time to start investing time and resources in things like segmentation, testing, trigger emails etc.

But there's more to the challenge than just improving targeting (which is what nearly every advanced tactic is about at its heart).

Consider why older addresses tend to disengage from your emails.

Some reasons are inevitable: people change interests, move jobs, etc. But there's also a parallel to population ecology. Um...what?

A population of animals can survive the occasional dip in numbers, provided the number of surviving animals never drops below a certain extinction threshold. If it does, the population is doomed.

I like to think of subscriber interest the same way. Your "relationship" with the subscriber can survive the occasional dip in interest, provided this interest doesn't drop below the extinction threshold.

In other words, you can send the occasional irrelevant email and people will forgive you. They'll still look out for your next email.

But if enough emails in a row are irrelevant, then the loss of interest dips below an extinction threshold. The recipient switches off and future emails can get ignored, even those emails that are valuable and relevant.

And the longer someone is on your list, the more chance there is that they'll encounter one of those catastrophic interest drops.

The key point here is that I don't believe that advanced targeting techniques are a complete answer to that disengagement problem.

A retailer who sends out offers cannot expect people to keep opening and clicking on promotional emails, however good and targeted those offers might be. I'm not going to buy a new digital camera (or camera accessory) every week or even every year.

See engagement as its own objective

If engagement becomes critical to delivery, then you must be less tolerant of letting people go inactive, even though many of these inactives may still be "unemotionally subscribed" and likely to buy/respond at some later date.

So how do we keep them engaged?

First, we can start to think of responses not just as sales, downloads, registrations etc., but also in terms of engagement. So what we used to think of as purely process metrics (opens, clicks etc.) might now become valid goals in themselves.

This encourages us to invest in email content that drives interaction: feedback surveys, teasers, etc. It also suggests those selling via email should also focus on engaging content as much as targeted offers.

For more on that concept, read Should all email marketers become content publishers?

And I wonder if we might reach the stage where the occasional email is used as a loss leader...where the offer is so good that it may even have a net financial cost to the sender to make it.

But the loss leader email boosts response. It keeps people away from the "extinction threshold" and keeps those engagement metrics up so that future emails have a better chance of being delivered and responded to.

What do you think?

Beware the wisdom of the minority

2009-09-03T13:36:00.000+02:00

In politics and society, we are rightly concerned with protecting minorities. It's a little more complicated in email marketing.

Email campaigns usually produce a nice set of numbers for us to enjoy: clickthrough rates, downloads, sales generated, opens/renders recorded, etc.

We're pressed for time, so we look at those broad numbers and decide whether the topic or offer in the email was a success or not. And we let that insight guide what we put in future emails.

Which is jolly good.

But there is a trap waiting for the unwary.

We assume the responses reflect the shared opinion of the entire list. But a vocal minority can disguise the true opinions of the majority. Which leads you to make poor decisions on future email approaches.

Let's explore three examples: misleading response boosts, viral effects and feedback.

Response jumps and segmentation

Assume the average clickthrough rate for a travel email list is 4.2%. Here are CTR results for the last four emails:

Clearly, emails 2 and 4 resonated best with our audience. We should consider modelling future emails on those two success stories.

Or should we?

What if we had two types of people on our list. All 100,000 subscribers are interested in travel offers, but 20,000 subscribers are cruise fanatics (Group A) and 80,000 are not really into cruises at all (Group B).

Let's now break down our four CTRs on a group basis:

Email 2 did indeed resonate better than average across the whole list.

But email 4 is a different story. The overall CTR was above average. But while Group A loved the message (perhaps it was a super cruise offer), Group B actually produced a worse-than-average response.

The strong approval from the smaller Group A more than makes up for the bigger group's disapproval. The result is an average that makes it look like email 4 is as broadly popular as email 2.

If we continued to send cruise-related emails to the list, we'd get above average results, not realizing that we're actually sending inappropriate material to the majority of subscribers. We're missing out on a huge opportunity.

This is why segmentation is so important: defining smaller groups with shared characteristics. So you can send email that has a better chance of a higher response and a lower chance of driving away others.

The trick, of course, is identifying those segments.

The wider lesson is that if you treat a list as an amorphous mass, you can end up sending emails that produce good results, yet are actually failing to engage a significant proportion of recipients.

It's one reason why large numbers of email addresses turn inactive on an otherwise successful email list.

Viral impacts

The articles you send out via email come with a nice "share on Twitter" icon. The number of tweets mentioning the article is a good measure of how valuable and "shareworthy" that article is.

But viral success is not just measured in total mentions. You need to look at who's sharing and the pattern of sharing, too.

In this model of sharing, your content has a broad appeal within your subscriber list, with many different people tweeting about it:

In this second model, the net outcome is the same, but there's a different sharing pattern.

Most subscribers were indifferent to the article. But it caught the attention of one subscriber who fed the link into a large community of like-minded individuals.

If that community is your target audience, great. But the chances are the article in Model 1 was actually more successful for you.

It resonated with a larger number of your existing audience and reached a more diverse group of individuals on Twitter. Individuals who are more likely to include potential subscribers.

Case in point: one of the most popular articles here at Email Marketing Reports is a collection of stats on email and webmail use.

It gets a lot of blog links, delicious tags and twitter mentions. But all of them come from email users and analysts, not email marketers.

The article is very shareworthy, but it's value to me is not nearly as much as, say, the article on HTML email design resources that gets less total "viral reach", but a more targeted one.

Feedback

Despite all the talk about engagement and dialog, most email campaigns do not produce a stream of return emails with feedback and commentary on the content or offer.

In fact, unsolicited feedback (or even solicited feedback) is often so rare that when it does arrive we give it far more attention than it deserves.

That's not to say feedback is irrelevant. Quite the contrary.

The chance to engage with a reader on a personal basis is a super opportunity to learn and impress. The danger is when you give that opinion too much weight in the wider scheme of things.

If that small amount of feedback is truly representative of the wider list's opinion, then you'll see that opinion reflected in your campaign numbers. Which is why it is always important never to look at a factor in isolation.

Vocal feedback can even reflect the actual opposite of broader list opinion.

Engaging emails with a strong personality often bring the best results. Yet a strong personality also carries a higher risk of offending or annoying some people for whom that personality is inappropriate.

As I wrote many years ago:

"It's quality, not quantity, that counts. If you lose 10% of your readership by changing your newsletter, but your impact and influence on the remaining 90% has improved tremendously, then the loss is a welcome one."

Related posts: Misinterpreting email marketing statistics

8 email statistics to use at parties

2009-08-28T16:28:00.000+02:00

Any self-respecting marketing channel is incomplete without a set of cool numbers to go with it: numbers you can print on a t-shirt and impress your friends with (if you have the right sort of friends).

Social media has them. Now it's email's turn...

If email was a country, its 1.4 billion users would make it the largest in the world. Bigger than China, bigger than the populations of the USA and European Union combined.

247 billion emails are sent each day. That's one email every 0.00000035 seconds.

In the time it takes you to read this sentence, some 20 million emails entered cyberspace.

Every second, the world's email users produce messages equivalent in size to over 16,000 copies of the Complete Works of Shakespeare (assuming a 30KB average email size).

13.4 billion: the number of direct marketing dollars forecast to go on email in the US in 2009.

$583 billion: the return from that investment if you use DMA figures on email marketing ROI. That's four times the market value of Microsoft.

181: the number of marketing emails it would take to produce enough revenue to buy one share in Microsoft.

83,689,738,832,367: the number of marketing emails it would take to produce enough revenue to pay the US National Debt.

[Aside 1: When you see email stats like the above, you gain a new appreciation for the work of those companies and organizations managing email.]

[Aside 2: And when you consider that many of these emails are spam, you can understand why ISPs and others have bigger problems to worry about than whether legitimate marketing email is reaching the right destination.]

Sources:
Email marketing ROI and other financial statistics
Number of email users and sent emails
Population of the EU
Population of the USA
Population of China
How many bytes in the CW of Shakespeare?
US National Debt
Market value of MS
MS share price

Video email: 8 recommended practices

2009-08-25T15:49:00.001+02:00

The first half of our look at video in email reviewed those scenarios where video makes sense as email content. Now let's get down to some specifics.

What tricks, traps, tactics and practices do our five video experts have for us?

1. Use video as a means, not an end

Wayne Wall, CEO at Flimp Media, a rich media marketing and analytics service, notes that your video is:

"...a means to get your audience to follow through on a call to action."

That's easily forgotten in all the excitement about "video experiences."

The actual business objectives of the email can suffer if, for example, the video "reveals all" without offering some way to take the next step toward a sale, download, registration, donation, etc.

2. Find the right video/copy balance in emails and landing pages

That means ensuring the video and surrounding copy complement each other. You need actionable links near the video, otherwise (as Wall puts it):

"...you give your audience no choice but to say, 'That was cool,' and delete the email."

It's a point echoed by Anna Yeaman of email design agency Style Campaign. She says the combination of video, text and images in the email mustn't do too good a job if, for example, you want to drive traffic to a website:

"For instance, we surmised that between the video gif and step-by-step recipe in this email, subscribers were left with little reason to clickthrough. Whether that's good or bad depends on your objective."

The trick is to find a balance between an engaging video and action-oriented copy. The latter should provide an outlet for those seeking to explore further, but without drawing too much attention away from the video itself.

Justin Foster, co-founder of video commerce solutions provider Liveclicker adds:

"Don't over-complicate your email with long copy or several different prominent calls to action when featuring video...people cannot view the video and read outside the video at the same time."

Similar principles apply to videos featured on landing pages. Chad White, Research Director at agency Smith-Harmon and author of the Retail Email Blog says you need to make it very clear what you want subscribers to do after they watch the video:

"Hallmark recently sent a back-to-school email that leveraged a TV commercial that they made. The point was to sell greeting cards that gave encouragement to students on their first day back at school. The email contained the video promo plus related cards, but the landing page didn't include any card promotions."

"In fact, it didn't include any Hallmark branding or navigation. It was just a video console on an otherwise blank page. That's a huge wasted opportunity to leverage the engagement of the video to sell products."

3. Don't forget video analytics

Obviously you'll be measuring and testing the impact of video content on the results of your email campaigns, through clicks, sales, forwards etc.

But you can (and should) also measure how people engage with the video itself. Flimp's Wall notes:

"...producing and distributing video is only half the battle. Once it's out there, you need to be able to monitor what it's doing."

Justin Foster agrees, suggesting you "...measure the engagement of video in email to learn how much of the video your subscribers actually watch."

How?

Well, your video email service may offer engagement analytics, where you can see, for example, how long people watch the video and when they break off to go elsewhere.

Anybody who has put up a video on YouTube will see a very basic example of this with YouTube's Hot Spots feature.

4. Keep it short and sweet and draw attention to the player

These engagement metrics will help you understand how long your videos should be. But Foster suggests 10-30 seconds worth of video in the actual email is a good target to aim for.

Yeaman adds:

"As a designer I get frustrated with the size restrictions of video gifs. Adding 10 seconds to the length will not impact performance so much as adding 50 pixels to the player width."

"For instance this Kraft food video is only 240x144 pixels. In order to offset the size, I added a drop shadow and orange box. Try adding design elements around the player to help it stand out."

5. Don't disguise the video link

Drawing attention to the video is particularly important where you use just a linked screenshot or still. This is where many marketers lose out, because they fail to make it clear that the image actually leads to a video.

Yeaman believes a static screenshot can hold its own against, for example, video gifs, but only if well executed...

"Pick a compelling still, place a play icon over the top, add a play bar, length of the video and repeat your CTA in HTML text."

...and she points us to examples from Express and Dillard's.

6. Design for failure

Unfortunately, even the best-designed screenshot can break down when images are blocked, which means alt attributes and adjacent text links are a must (for more on coping with image blocking see here.)

And even the best "video in the email" technology cannot guarantee full functionality in every situation. So you need to design for when the video format is not supported.

Peter C. Horan, Chairman and CEO of Goodmail Systems notes that their CertifiedVideo technology has built-in redundancy, where the video:

"...falls back to today's 'static video image with the click arrow' should a consumer's browser fail to cooperate."

Video gifs are image-based animations so face image blocking constraints, too. In addition, the Outlook 2007 email client only displays the first frame of such an animation. So it's worth giving creative thought to how that first frame looks.

Anna Yeaman says:

"One technique we've been using is to add a play icon over the first frame for Outlook 2007 users. You might take that a step further, by replacing the default first frame with an engaging still."

Justin Foster also points out that different clients vary in terms of optimal frame rates for video gifs.

"Video .gifs work 'nearly' everywhere, as opposed to 'everywhere.' For example, Outlook 2007 will not display video. Some mobile clients will show choppy video, while others support only static images."

"Video .gif delivery technology can help compensate for some of the client-specific limitations by automatically swapping in static images or varying the video frame rate based on the detected client."

The lack of audio in video gifs also demands creative thinking. You can't simply repackage an existing video that has sound, if the audio is an important part of the message.

7. Test your options

With video email there is a tradeoff between the quality of the in-email video experience and the cost. Whether you go for a video gif or CertifiedVideo, it's worth testing against the (cheaper) simple screenshot linked to a video hosted online.

Chad White's company, for example, "...A/B tested a video gif vs. a standard video image tease. The email with the video gif dramatically outperformed the one that just used a static image to promote the video content."

Justin Foster also notes that "...a 5%-25% increase in sales has been demonstrated in A/B tests featuring in-email video v. the 'gold standard' of video in email (using a static image with a play button overlaid, linked to a landing page featuring full video)."

It's important, however, not just to compare the performances of the alternatives in terms of your email goals, but to balance that against costs.

8. Be aware of production and bandwidth costs

Anna Yeaman warns that for smaller organizations those costs:

"...can outweigh the benefits. Video production fees, editing, converting to gif, delivery and or bandwidth charges are currently a deterrent for many."

You have some cost flexibility with regard to content generation. Wayne Wall says a video doesn't have to be expensive to be effective:

"The prevailing theory is that video takes lots of time and effort to get it 'right.' But there's no 'right' way to do video. We've seen effective campaigns shot in a half hour on $100 flip cameras."

Nevertheless, delivery and bandwidth can incur significant costs, depending on the technology and approach you use.

Justin Foster notes:

"It's quite possible for video email to consume anywhere from 10X to 500X the amount of bandwidth required to send 'normal' email...email marketers used to sending 'on the cheap' could be in for a case of sticker shock."

The actual physical infrastructure required to feed video files to a large list is also an issue. As Yeaman says:

"If you're sending out to a small list, your ESP might accommodate you. For larger lists you'll need to partner with a video vendor, or find someone capable of hosting the video file."

Foster also recommends using dedicated content delivery services to ensure videos display smoothly regardless of the email subscriber's physical location.

The technical nuances of videos playing in the email itself suggests the technology may not be appropriate for a DIY approach yet unless you're sure of your expertise and infrastructure.

Video email: when to use it

2009-08-18T14:32:00.002+02:00

A year ago you might have asked "can we put videos into emails?"

Today, the question is more sophisticated: should we use video and, if we should, how?

To provide some answers, I picked the brains of five video email experts representing a mix of technologies and perspectives.

Before we see their insights, here's a quick review of the main video email approaches out there (see the video email category for details):

Link a screenshot to a video hosted at a website (more info)
Video gifs: optimized animations that play like a video in the email itself, albeit without audio (more info)
CertifiedVideo: full rich-media audio and video content that also plays in the email itself, currently available at AOL only but with plans to expand to other ISPs (more info)

Today we'll highlight some promising approaches and good examples. Next week we'll look at specific advice on implementation and discuss some of the constraints and mistakes that can limit the tactic's usefulness and success.

When should you add video content to your emails?

For a long time, a simple justification for video content in email was the novelty factor: moving pictures in an email attract attention and pique curiosity.

While this is still true to some extent, today's video content - as with all email content - needs to offer something more if it's to work effectively.

"New toy syndrome" is a danger if marketers use video email because they can and not because it offers a true benefit to both sender and recipient.

Justin Foster, co-founder of video commerce solutions provider Liveclicker warns:

"Done poorly, video in email can be a distraction, so email marketers should always ask the question: is this video actually adding something important to my message, or is it just annoying?"

It's a point reinforced by Peter C. Horan, Chairman and CEO of Goodmail Systems (home of CertifiedVideo), who says his company advises against using video email...

"...when it's video for video's sake and doesn't really benefit the consumer."

And, as with all content, it's better to have no video than bad video. Horan says:

"There is absolutely nothing worse than boring content, broken links, static images and slow video streams."

Use video when it offers more than just a novel alternative

Video's primary value then comes not as a straightforward alternative or complement to text and images, but where the video itself offers a better experience than text and images can.

As Anna Yeaman of email design agency Style Campaign puts it, video is particularly apt when the equivalent copy and images would be "too lengthy or inadequate" for the job in hand.

It's a theme taken up by Chad White, Research Director at agency Smith-Harmon and author of the Retail Email Blog:

"Video is fantastic for things like product demonstrations, how-to information and conveying other information that's tricky to do with words and images alone."

Both Yeaman and White highlight Apple's email launching the iPhone 3G as a good example. As White says:

"For a slick, tactile product like the iPhone, showing a hands-on demonstration of its capabilities is more powerful than just listing out capabilities."

But the speed of information transfer via video means that where words can do the same job, they perhaps should do the job. White explains:

"One of video's drawbacks is that it's slow. It controls the pace at which you're absorbing the information. So if the information can be conveyed just as easily in writing, then you're probably better off going in that direction because people can read faster than someone can talk at them."

He warns:

"Make sure you're using video to convey the right content or hurried subscribers will tune you out."

Three main models

Goodmail's Peter Horan notes that early adopters of CertifiedVideo have tended to follow one of three directions, using video where:

1. It's the primary product, such as news and entertainment. Horan gives an example:

"You're promoting the Jonas Brothers next tour to teenage girls. Targeted email with "teasing" streaming video will increase interaction, prompting the viewer to watch a longer clip..."

2. It can demonstrate how a product works, similar to the Apple approach illustrated above.

3. It can help to add an emotional overlay to a rational message.

Smith-Harmon's White mentions a Williams-Sonoma email promoting wreaths with a secondary call-to-action to See How They Are Made. The email demonstrates how video helps build in more engagement and a broader experience with a product...

"That video included a tour of the farm where the herbs and other wreath-making materials are grown, an interview with the owner of the family-owned farm, and footage of workers putting the wreaths together by hand."

"Williams-Sonoma had a great story to tell about how they're supporting a family-owned farm and craftsmen. The story added to the allure of the product."

What does the video medium say about what you want to say?

Wayne Wall, CEO at Flimp Media, a rich media marketing and analytics service, suggests exploiting the implication that a video format represents an important message:

"...if you have big news to announce, an audience will understand that it is important before they even hear the news by virtue of the fact that you're using a high-quality medium to engage them."

Wall also warns that video isn't excluded from the kind of best practices you apply to any email content, such as targeting.

It might seem self-evident, but the excitement over video email can lead marketers to forget some of the basics in the rush to play with a new toy. Wall says:

"You want your message to be well-received, and knowing how your audience responds to certain kinds of content will help to determine when video email is appropriate."

"A deathly serious video will fall on deaf ears if you send it to a younger audience; likewise, a fun, casual video might send the wrong message about your company to certain C-level executives."

In Part 2, our five experts explore some specific recommended practices for implementing video email and look at the constraints and bad practices that can turn your message into a horror movie.

More on videos in email | Tags: video emails, video gifs, goodmail

Subject lines: Amazon's lessons on branding

2009-08-13T16:06:00.001+02:00

A review of 187 promotional emails from Amazon reveals many insights about subject line length and personalization. Now let's see what these emails teach us about subject line branding...

[Please take the poll at the end of this post to decide if I should do more in this series.]

One important element driving reader attention and response is recognition. If the recipient doesn't quickly grasp the source and purpose of the email, she may simply gloss over or delete the message without much thought.

An obvious "from" header is key here, but many senders put their brand or business name in the subject line to reinforce this recognition. This post discusses the topic in detail.

For example, Amazon.com almost always puts their name at the front of the subject line. Amazon.co.uk, however, almost always puts it at the end. And Amazon.de sometimes puts it at the front, sometimes at the back, but mostly never mentions it at all:

Yikes.

Everything you put in a subject line has a benefit and a cost.

Each word (hopefully) influences the reader to pay closer attention. But each word carries an opportunity cost: it takes up space other words might use.

Faced by the numerous things we could put in the subject, we try and pick a combination of words that causes the best overall response.

The right combination depends not only on basic subject line common sense, but on context. Amazon's subject line branding looks like a classic example.

Branded subject lines are important in inboxes where attention is scarce, perhaps through email overload or large amounts of spam. Conversely, if someone gets email from Amazon and nobody else, branding the subject line would be pointless.

Equally, the stronger the brand, the more impact it likely has in grabbing attention and encouraging interest.

So try this theory...if we rank the US, UK and Germany in terms of Amazon's likely brand strength we get this:

Amazon.com > Amazon.co.uk > Amazon.de

And if we order their respective countries in terms of the likely total volume of marketing email received by typical email users (busiest inbox first):

US > UK > Germany

Now the different subject line branding strategies make sense.

Amazon.com uses its strong name to stand out in cluttered US inboxes.

Amazon.co.uk has some benefits as a subject line element, but not enough to justify priority placing, so the name gets shunted to the back.

In Germany's less-cluttered inboxes, Amazon.de saves the space for shorter, cleaner subject line phrasing.

The importance of context is also clear in the exceptions to Amazon's broader subject line branding approach.

Amazon.co.uk, for example, uses the brand name at the beginning of a subject line when the email's attractiveness is relatively low.

Compare, for example, these two subject lines:

The first one is a relatively weak promotion. In this case, the power of the brand to attract inbox interest is probably bigger than the power of the promotional message itself. So the Amazon.co.uk gets placed first.

The second one has a strong appeal with its heavy discount. There is more benefit to be gained from putting the discount up front than through branding. So the Amazon.co.uk gets shunted to the back.

Makes sense, no?

Someone publishing the same kind of email on a regular basis can see subject line branding as a yes/no issue. But others, as the Amazon examples show, may choose to tackle the issue separately for each type of email they send.

Again, see this post for more ideas on subject line branding.

OK, I have more insights based on Amazon's emails, but three posts on their subject lines seems like a lot already so here's a quick poll to see if you want more:

Do you want more insights from Amazon's subject lines(polls)

(If the poll won't display in your reader, click here.)

If the answer is yes, I'll take a short break for something different anyway, and return to the topic later this month.

More on subject lines | Tags: email marketing, subject lines

Subject lines: Amazon's lessons on personalization

2009-08-10T10:18:00.001+02:00

The last post used a review of 187 promotional emails from Amazon.com, Amazon.co.uk and Amazon.de to draw out insights on subject line length. But that's not all we can learn from the retailer's email expertise.

So today we look at what Amazon might teach us about personalizing subjects: is adding the recipient's name to the subject line a good or bad thing to do?

Since 11 of Amazon's 187 emails (5.9%) to me feature my name, the answer would seem to be "sometimes, but not often".

The various pros and cons of this kind of personalization are outlined in an earlier post, where the consensus suggests it's largely ineffectual and potentially counterproductive.

But if that was always true, Amazon wouldn't use it ever. So let's dig a little deeper.

Format

First, Amazon always uses my full name: Mark Brownlow, not Mark. Here's an example:

Mark Brownlow: Up to 75% off DVDs

One reason may simply be that they don't have separate database fields for my first and last names.

But does it have a psychological justification?

One argument against personalization is that spammers often use the same technique.

They take the bit before the @ in an email address and use that as the name. The result is often laughably inappropriate: "Hey info, buy our stuff".

So if you use the recipient's name, it might trigger a "this is spam" reaction.

But no spammer can pull out a properly formatted full name from an email address. So using the full name gives the personalization more credibility than a simple first name might have.

Additionally, the use of first name implies a level of familiarity which may not be appropriate for the way Amazon thinks I feel about the brand and customer relationship.

Already we begin to see that superficially simple choices like "name or no name" and "first name or full name" are more complex than we might think.

This kind of choice is one that each sender must consider in the context of their own business. There are no rules set in stone and your main refuge is testing to find what works best for you.

Geographical segmentation

A closer look reveals that Amazon.com and Amazon.co.uk hardly ever use my name in the subject line. Just once each in over 5 months' worth of emails, which were more than likely tests.

In contrast, Amazon.de uses name personalization in the subjects of around 17% of the emails they send me.

Why would they use my name, while their counterparts elsewhere choose not to?

Different cultures likely respond differently to seeing their name in a subject line. Since German email users are subject to less spam and marketing email per se, they may see less negative association with such personalization than in more spam-weary countries like the USA.

So in Germany, maybe the positive effects associated with seeing your name are not overwhelmed by negatives such as the potential association with spam.

At least one study confirms this with a German ESP reporting that such personalization was indeed associated with a strong boost to open rates (though they suggest the novelty factor helped).

User segmentation

It may not be cultural segmentation, but user segmentation at work. Digging even deeper shows that Amazon.de uses my name only on one particular kind of email: those announcing broad discounts on electronics.

This may be coincidence, but it shows the importance of viewing your list as a series of individuals or segments, rather than one amorphous mass.

Perhaps I responded well to a personalization test on a previous electronics mail?

Perhaps gadget freaks are more vain than others and respond more positively to seeing their name in the inbox?

There are also suggestions that the use of personal data like names works best when justified by the contents of the email. If the email is clearly customized to the individual, then personal data has a positive effect on response.

However, if an email is pitched as personalized (as implied by the use of a full name in the subject line) but the contents are clearly generic, then this creates a gap between the expectation set by that subject line and the value delivered in the email. Which leads to lower responses.

An email about electronics (which I buy a lot of), perhaps with product categories highlighted that I commonly browse or purchase, is tailored to my interests in a way that broader sales announcements are not.

So the use of my name might enhance and complement the personalized impression.

Highly-targeted recommendations

Let's assume we're right about personal data (like a name) helping when the contents are also personalized (targeted) to the recipient.

Surely, then, my name should be used in subject lines recommending cross-sells and upsells related to my previous purchases? It's hard to get more personalized targeting than that.

Yet none of the many cross-sell emails Amazon sent me include my name in the subject line.

Why not?

Subject line design is about getting maximum value out of a limited amount of space. A name has a benefit in some circumstances, but it takes up space...potentially diminishing the impact of other parts of the subject.

Personal recommendations like this one...

Amazon.co.uk recommends "Kung Fu Panda [DVD] [2008]" and more

...are already very targeted. The tie up with my previous purchases already implies customized content. The addition of my name is perhaps unnecessary to make that point. And subject lines for personal product recommendations are lengthy enough without adding another dozen characters for a name.

Novelty value?

Another reason why Amazon might use subject line personalization sparingly is perhaps to preserve the novelty value mentioned earlier.

The company sends me a relatively large amount of promotional email and consistent subject line formats can become monotonous. Perhaps Amazon likes to break up this monotony now and then through personalization. Just to stop me switching off.

Though we can't know exactly why Amazon takes the personalization approach it does, it's clear from the above that decisions about subject line elements need careful thought.

Which brings me to the next post in this review, which will look at lessons from Amazon's subject line branding and copywriting.

More on subject lines | Tags: email marketing, subject lines

Subject lines: Amazon's lessons on length

2009-08-05T21:27:00.002+02:00

I once interviewed a CEO who described his online retail philosophy as WAD-based. Whatever Amazon Does.

So what might WAD teach us about email subject lines?

Here's Part 1 of an analysis of the last 187 promotional emails I received from Amazon.com, Amazon.co.uk and Amazon.de.

Length

For the stat fans, the average subject line length used by Amazon was 57 characters.

But before we all carve that number in stone as something to aim for, consider the graph below. It charts the number of times a particular length of subject occurred in those 187 emails:

Blimey.

While the average is 57 characters, most subjects are actually shorter than that: between 39 and 52 characters. And there are still plenty of examples that are much longer than that average. In fact, the longest comes in at 140 characters! (The shortest is just 19).

So what's going on?

Many marketers assume that shorter is better for subject lines. That's not entirely true. The best subject line length is one that uses as few words as possible to provide the information or incentive that gets the reader to open it.

That's a little more nuanced than simply striving for 50 characters or less.

So, yes, keep it short. But not at the expense of important words that can contribute significantly to higher responses. (For more detailed discussion of this, see Part 1 of last year's subject line series and the scrabble theory post.)

Amazon is taking this approach, adjusting the length of the subject line to need. A simple email with offers on TV boxed sets looks like this:

TV DVDs from £4.97 at Amazon.co.uk (34 characters)

It's short and snappy.

But a specific book recommendation is longer, ensuring that the words likely to trigger my interest (the book's title and/or author) are included:

Save 22% at Amazon.com on "Speak: 10th Anniversary Edition" by Laurie Halse Anderson (84 characters)

Those specific product offers are based on my previous purchases or where I spent time on the various Amazon websites. So they're highly targeted in theory (a topic for another day).

The problem with such lengthy subjects, though, is that a lot of them get truncated by the various email systems out there. My Windows Live Hotmail interface, for example, only displays 29 characters the way I have it setup.

Which is why the very important words need to be at the start.

Consider:

Jetzt neu: "Fantastic Beasts and Where to Find Them: Comic Relief Edition" von Joanne K. Rowling

The new book announcement (which is what "jetzt neu" implies in German) is fine in theory. I've bought enough Harry Potter books through Amazon Germany, so I'm likely interested in anything new from J.K.Rowling. Super targeting!

Here's how it looks in my Thunderbird inbox:

The targeting value is lost in this example, because the hottest "trigger" words are actually the author's name. Which does not appear.

You might think it's a simple matter of reversing the sequence of data pulled from the database: author - title, rather than title - author.

The problem is that the relative importance of a product's name, function and brand is not always the same.

In the above example, the author's name was critical. So reversing the order would work well. But in this next example, it's the title that's critical. The product is a sequel to the bestselling Gruffalo book:

Jetzt neu: "Gruffalo's Child (Gruffalos Child Big Book)" von Julia Donaldson

In this case, a truncated subject line is no problem as the key information is still visible.

Reversing the order to author - title would hurt the targeting (with all due respect to Julia Donaldson).

There is no simple answer to the problem since it relates to automation and product database fields, for which I have less expertise than my cat.

Perhaps those looking to integrate email with customer purchases and browsing behavior can have a single product field in their database specifically for use in email subject lines.

So our first example might look like this in an email:

Jetzt neu: J.K.Rowling's "Fantastic Beasts and Where to Find Them: Comic Relief Edition"

And the second like this:

Jetzt neu: "Gruffalo's Child (Big Book)" von Julia Donaldson

Perhaps some retail email experts out there can offer more advice on dealing with this issue.

Amazon's longest subject line illustrates another problem with automated emails pulling data from a product database:

Save 25% on "A Cook's Book of Sauces: Mayonnaise Hollandaise Bearnaise: Mayonnaise, Hollandaise, Bearnaise (Murdoch Books)" by Murdoch Books

You can see a similar problem with word repetition here:

Save 26% on "Biker Mice From Mars - New Cats In Town [DVD]" on DVD

So when people talk about triggered emails etc., be aware that it's not a perfect technology!

OK, next week we'll look at some more Amazon-inspired subject line insights on personalization, branding etc.

More on subject lines | Tags: email marketing, subject lines

Email and Twitter: more observations

2009-07-24T13:30:00.002+02:00

Since I use email and Twitter to promote this site, does that make me a Twailer? Or an Emitter?

Anyway...as I'm on vacation, it seems a good time for reflection and this post outlines brief observations on some important differences between the two channels and what these differences might mean for your marketing.

I've outlined how the two shape up against various criteria, like response, relationship building etc. and end with a summary of insights.

This is very much one man's perspective so do chime in with your own opinion in the comments. There's also a complementary post with earlier observations on the Twitter Adventure.

Response rates

A recent article at Mashable suggested an average CTR on links in tweets of 2.8%.

Before we all rush to gloat about email CTRs, a direct comparison is unfair.

Links in marketing emails are intended to be clickworthy. They (hopefully) have a clear marketing purpose, clear value to the recipient and a clear call to action.

Links in Tweets may also share that quality.

But the nature of the channel also encourages you to send people links with no direct-response business objective in mind. And those kinds of links can be very hit and miss.

Throw in that it's much harder to target in Twitter and we'd expect average responses to be less than for marketing email. But...

1. Heavyweight links can also get good responses at Twitter
2. The Twitter environment is more forgiving of multiple tweets on the same topic (featuring the same link)
2. Retweeting (the Twitter equivalent of a forward) can have a big impact on response

So for "standard" content, response rates for email are broadly better than Twitter. Twitter has far too many distractions to let people respond to the mundane.

But for "standout" content that is shareworthy, Twitter can be better than email. Retweeting can drive a message around the world in minutes, since Twitter effectively supports one-to-many forwarding...email largely just one-to-one forwarding (notwithstanding all our efforts to add Share With Your Network links to messages).

Of course, there is also the issue of scale and reach (see later).

Relationship building and dialog

Twitter forces you to confront the fact that the people listening are...um...people. With email, there is a tendency to treat email addresses as numbers (it doesn't have to be that way, but it often is).

As a more personal environment, Twitter allows more direct, personal relationships than email. But a key rule applies in both channels: send a lot of meaningless, self-absorbed, self-centered messages and you'll fail in both environments. Only your true friends will keep listening and few of your customers are true friends.

Since Twitter is both personal and public, it encourages dialog and participation by a greater number of people. Assuming you're delivering value to the partners in that conversation, this also helps with relationship building.

Email builds relationships and dialog in a more subtle way. All the targeting tools at our disposal let us deliver value, which encourages loyalty. And the modern obsession with brevity neglects the fact that some information is better expressed or exchanged in more than 140 characters.

Analytics and subscriber data

The nature of Twitter makes it hard to build out data on individual followers. And even if you could, it's impractical to target content to segmented groups of followers at any scale.

That's where email scores well. You may already have detailed information on the subscriber if the sign-up comes out of a transaction. Equally, you can build up information on subscribers based on how they interact with your emails.

However, Twitter has the advantage that you can normally put a name, face and website to each follower. This is particularly important to those interested in building one-to-one relationships with a small group of individuals.

You can also learn a lot about your audience via Twitter, if you take the time to listen. Excel won't help you much so you need softer, more intuitive analytical skills.

Costs

Twitter messages don't need coding and design testing. They don't break in Outlook. So message production is cheaper and easier than email.

But Twitter's advantages in dialog and relationship building carry a cost in time. It encourages, thrives on and demands immediate responses and interaction which can place a heavy burden on your time.

Subscriber acquisition

Reach and scale is a big issue. Twitter is marvelous at building dialog and relationships, spreading shareworthy content and gathering intuitive intelligence on your audience. But its reach is still small. Any one of the big four webmail services alone manages more accounts than Twitter.

Low costs per tweet turn into high costs per click if you reach ten people rather than 10 million. Email on the other hand is ubiquitous.

Where Twitter scores well is that people will subscribe to (i.e. follow) hundreds of Twitter accounts, but only subscribe to a handful of email lists. That's a double-edged sword. While it may be easier to get people to follow you (always assuming they use Twitter), it also means your messages are competing with hundreds of tweets.

Risk

One comparison of the two channels rarely mentioned is risk.

Email marketing is not hostage to any one delivery service. You need not be reliant on a third-party service at all if you choose to use in-house software. Equally, if your email service provider declares bankruptcy tomorrow, you could switch to another one relatively quickly (not painlessly, but relatively quickly).

Twitter is Twitter. If the company ceases to operate, it's over. You can start again at another microblogging service and hope your Twitter followers migrate with you. But the fact that I can't even name an alternative service bodes badly for that. You would have to hope that former Twitter followers switch to another channel to hear from you.

Closing insights

None of the above should suggest that Twitter and email are somehow competitors. Even these quick observations make it clear that they are different tools suited to different purposes.

Email people who reject Twitter are missing an opportunity. Twitter people who reject email are missing an opportunity. No channel has some inherent superiority: judging either depends on what it can do for your business given your goals, audience and resources.

The time aspect is critical here. Jay Baer's quote on social media thought leadership is apt:

"You have to set social media limits, or you'll drive yourself crazy trying to be everywhere at once."

Both channels need the proper investment in time and/or resources to bring the best rewards. Both channels work for marketing when you recognize that those on the receiving end of the messages are looking for value from you. Attention comes with expectations and fulfilling those expectations takes commitment.

Thoughts?

Improved email marketing search engine

2009-07-21T14:35:00.000+02:00

Just added 35+ new sites to the database for the custom email marketing search engine powered by Google at OnlineMarketing.info.

You can now search over 360+ hand-picked sites to uncover email marketing info without having to wade through any spammer-targeted rubbish that might have successfully gamed the traditional search listings.

Latest additions include blogs like Spamtacular, Convince and Convert and Red Pill Email, email galleries like Beautiful Email Newsletters, tools like Litmus and various services.

If you feel a site deserves listing in the search engine, leave a comment and I'll check it out on my return from vacation.

Identifying engaged subscribers: unique opens, clicks, lateral thinking

2009-07-17T14:25:00.000+02:00

Last week we saw why it's important to identify your engaged or "best" subscribers. And we discussed how repeat opens is one way of doing this when you don't have funky customer data at your disposal.

But what other techniques can we use?

In this post, we'll cover unique opens, clickers and how we might apply lateral thinking to come up with other neat measures that don't require much work or data.

Unique opens

Your average subscriber does not open every email from you. So if you're looking for engaged subscribers, how about searching for those who open more than the average?

As in the previous post, I took four months of emails (nine issues of my e-newsletter) and used simple campaign reports to evaluate open patterns. About 68% of the list recorded at least one open over that period.

Of those 68%, this graph shows how many opened one, two, three etc. of those emails:

We see that 35% of our openers opened at least half of those emails. And 8% opened every single one. So we might consider those 8% to be our best subscribers.

There are, of course, two problems with this approach.

First, a single open does not imply anyone actually "engaged" with the email, which is why we looked at repeat opens last time. A recipient might click on every email, trigger an open in a preview pane and delete it without reading.

That would put them in the 8% but you'd hardly call them engaged.

Equally, there are those reading the emails religiously who never trigger an open because they use an email client where images are blocked.

Fortunately, even basic email campaign reports also tell you who clicks on an email. So instead of looking at unique opens, perhaps we should look at unique clicks. After all, if someone clicked on a link then they must be interacting with the email.

Clickers

A look at the campaign stats for that B2B newsletter of mine found about 40% of the list had clicked on at least one of those nine emails.

So here's the same graph, again. This time it records how many of those 40% clicked on one, two, three etc. of those emails:

If you compare the two graphs, you see why it's so important to look at numbers that represent genuine interaction. About 25% of the entire list opened at least half of those nine emails. But only 3.5% of the entire list actively clicked on over half of those emails.

Still, we can treat those 3.5% as genuinely engaged subscribers.

Apply lateral thinking

If these open and click patterns aren't enough, how about digging deeper?

Example 1: If someone really likes your emails, they would maybe keep them and refer back to them later.

How about looking at open times or click times for past emails and isolating those subscribers who opened or clicked weeks and months after the email went out?

Example 2: We know that the longer someone is on a list, the less likely they are to open and click on emails.

So someone who opens every second email and signed up in 1997 is likely far more engaged that someone who opens every second email and signed up last month.

So how about splitting subscribers into segments based on sign-up date, calculating average engagement metrics for each segment and then picking out those who engage well above-average in each segment?

Example 3: Are all clicks equal? Are there some links that suggest only those with a close relationship to your brand would ever bother clicking? Are there links where you can immediately tag anyone clicking them as a "top" subscriber?

What about "behind-the-scenes" links like "learn more about the team" or "see how our factory works"?

What about people who click on customer survey links and complete the survey without any kind of incentive?

Can you find links that identify engaged subscribers in that subtle way.

My newsletter has the title "No man is an iland" and an obscure link nearby that reads "Why no S?":

I might decide that anybody wanting to know the origins of the newsletter's title is likely more than averagely interested in what I have to say. Incidentally, 2.66% of the list have followed that link.

What other measures can you find to "out" those best subscribers?

The ultimate subscriber

If you're not happy with one or another measure, you can use two or three, compare the addresses identified and see if there's any overlap. Anyone who comes up as "engaged" using each of several different measures really is a "best subscriber".

I did the exercise and found one subscriber who fits all these criteria:

Actively subscribed for over two years
Clicked on more than half of the last nine emails
Registered an open on each of the last nine emails
Clicked at least once on the "Why no S?" link
Recorded more than five times the average number of total opens on at least one email
Recorded more than twice the average number of total opens on at least two emails

Wow. I think they deserve a reward, don't you?

(And, no, it wasn't me.)

Final thought

Many of these measures of engagement are biased by those who signed up sometime over the period of comparison.

Sometimes these new subscribers are unfairly penalized: it's hard to click on more than one email when you've only ever received one email.

Equally, new subscribers tend to open and click more often than old timers. So they can appear more engaged than they truly are.

This high level of initial engagement is why it pays to send new subscribers a specific stream of welcome messages that take account of this factor.

It may also make sense to keep new subscribers out of your "top subscribers" pot until they prove their continued interest after the initial novelty of your emails wears off.