Metrics Posts

After the Click: Improving Campaign Performance With Web Analytics

Written on April 19th, 2010 | Posted by Marco Marini in Email, Email marketing, Analytics, Metrics

What happens after you send out that carefully crafted email campaign? What do you look for beyond open rates and click throughs? What do people actually do when they get to your landing page or website? And are you able to not only track that information, but put it to use in your next campaign?

To really understand the effectiveness of your email marketing campaigns, you need to keep tracking your customers’ behavior beyond your email, even beyond your landing page. How else can you know your real success if you don’t know your real results? Plus you can learn from knowing what people do at your website. Where else do they click? Which other pages do they visit? Do they sign up for your emails? How long do they hang around? At what point do they bail? To gather this data, your email must be integrated with your web analytics. It’s the only way to carry your tracking through comprehensively.

And on the other side of that information gathering is what you do with that data to improve your email marketing. Ideally it’s a closed loop process, with the ESP and web analytics both feeding information to each other.

By integrating your email with your web analytics, you can track behavior and better understand your conversion rates, improve your campaigns, respond to individual behavior in near real-time, and ultimately increase your email marketing ROI. You can learn, tweak and improve, and even segment your email marketing messages in the future.

Integrating email with web analytics gives you real-life data, but it’s not as easy as it seems. However, the payoff is worth it. If you’re ready to take on—and profit from—this kind of integration, here are some things to consider, both when choosing an analytics provider and when setting up the integration:

  • How often data is flowing from the analytics provider to your ESP and how quickly do you need to make decisions based on that data? If you can wait 24 hours to get data back, then a batch process is fine. However, if you’re looking at shopping cart abandonment, and you need to react right away to a behavior, you need something more real-time so you’ll want an inline process that allows immediate reaction, without the delay of a batch process.
  • What segments are important, and what information do you need in order to allocate or define the segments?
  • What is your internal availability for building an API now plus supporting it later? Do you want your IT team to take this on, working with your marketing team? Do they know the ESP well enough, and can they support the integration when something goes wrong? Or should you outsource this?
  • How easy is it to migrate if you switch ESPs? You have to make sure your new ESP can tag links to where your analytics package can easily identify the same information for a person, and for a campaign. That new ESP is also going to need to be able to consume data from the analytics company, send and consume data back and forth from the analytics package.
  • As a preventive measure, your marketing department needs to acid test the solution. You have two separate systems operating relatively independent of each other, but you need to regularly make sure the information going back into both systems is accurate.

When your email and web analytics are integrated as a closed-loop process, it should be seamless. Your ESP sends an email and links within the email include identifiers of who that recipient is and the campaign they’re being sent. When they click on a link, the ESP feeds that information over to the analytics provider on a batch basis. You’ll learn about the campaign performance, but also specific metrics about who did what individually once on the site. But that’s not the end of it. The web analytics can also feed information back to your ESP, enabling automated responses or other email messages appropriate to a customer’s particular actions. Then you take all you’ve learned and tweak your next campaign accordingly.

If you’re tracking email and web analytics separately, you’re missing the big picture. If you have them integrated, you’re ahead of the game…and the competition. 

- Marco Marini
ClickMail Marketing



more » | comment »

A Call to Action for Standard Email Metrics

Written on January 31st, 2010 | Posted by Stephanie Miller in Eec, Statistics, Metrics


The email marketing industry needs standard reporting and metrics.

Today it is impossible to compare and benchmark response and deliverability rates across the industry because marketers get reports with different terms based on different calculations. Marketers are restricted in comparing reports and synchronizing data when looking to evaluate or change email broadcast vendors.

Inaccurate or inconsistent metrics diffuse the credibility of email marketers.  If our own metrics cannot conform to benchmarks, we lessen our ability to convince senior management and fellow digital marketers of our success.  It also hinders our ability to negotiate for resources.

You can help.  Read the quick background here and then take action with the links below.

The email marketing industry may be ignobly unique among direct and online marketing disciplines for our lack of measurement standardization.  For the past two years, the members of the eec Measurement Accuracy Roundtable (a volunteer member committee)  have wrestled with the problem of a lack of a consistent and unified standards for the most basic email metrics such as delivered, open and click.

Through our work, the Roundtable has built a foundation for industry standardization for these basic but important metrics.

We have created (and vetted) new definitions of key measures so that they are not only accurate, but the names accurately reflect the measure.  (You can read in past eec blog postings about the struggles and debates to come up with terms we could all support.)  Latest definitions are here.

We have surveyed dozens of email broadcast vendors (ESP's and MTA/on-premise providers) in order to audit existing reporting and gauge the level of variance across the industry.  Please note that the eec Roundtable does not support or claim that any one provider’s method of calculating common metrics is better than any other.  Many ESP's and other broadcast vendors participated in the development of these definitions.  We are very grateful for their support.

The Roundtable has repeatedly come to the industry – practitioners, eec members and thought leaders – to gather feedback and insights.

Now it’s time for action.

Here’s how you can help us start the ball rolling.  Join our launch efforts now.

Voice your support (or dissent) for standardization of metrics in our industry.  Take this one question survey.

Read the definitions

Tell us your thoughts and send in any corrections to the Roundtable.

CommitSign the petition to advance standard metrics now.

Join the Roundtable (eec members only).  Just email Ali at the eec.

Please place your comments below.  And stay tuned!

Thanks to the hard working members of the eec Measurement Accuracy Roundtable! 

- John Caldwell & Luke Glasner, eec Measurement Accuracy Roundtable Co-Chairs



more » | comment »

When It Comes to Metrics, It’s All About You

Written on April 6th, 2009 | Posted by admin in Deliverability, Metrics

As the first quarter of the year recently ended, I would like to take this opportunity to discuss a question that I keep getting asked by current and prospective clients: “What is the industry average for delivery rates?” As much as I’ve talked about this with clients and in public forums, the question still comes up – and my answer is always the same: most marketers should not spend their time being concerned with industry averages for email delivery.

Here’s the problem. When marketers focus on what others in the industry are achieving, they are spending less time focusing on their own programs. To me, it shouldn’t matter what others are achieving; it matters what you’re achieving.

If the industry is getting an 88% delivery rate, but you are getting a 90% delivery rate, does that mean that you should stop trying to improve it? Instead of sitting back and enjoying the fact that you are getting 2% more of your mail to your users, it would serve you better to try to increase that delivery rate by another couple of percentage points and increase your ROI.

Of course delivery rates aren’t the only things marketers focus on; opens and clicks also get a lot of attention. The standardization of these numbers is important for the industry, and especially for senders — not only for properly evaluating the services of a prospective provider, but also for making sense of all the studies that discuss these numbers.

For example, say a company comes out with a study that finds that for industry “x”, the average delivery rate is 85% and open rate is 5%. Without standardized metrics, how do we know if all of the companies that were surveyed actually determine their numbers the same way? What if the open rate for one company was determined by number of messages sent, while another was calculated by number of messages delivered? Depending on their independent delivery rates, this discrepancy could have a huge effect on the reported results.

As you look at all your numbers, remember that while it’s good to know what others are doing, it’s more important to track how you are doing and continue to improve — no matter what your numbers are.

Here’s the other thing to remember when reviewing various reports, surveys and other data points that may or may not make you feel good about your programs: You can never control what someone else does; you only have control over your own actions. That means, as always, report on this information for your own company, or perhaps your own division. Then watch it, track it and use it.

Are you seeing the same thing with your clients? Let me know in the comments, along with any other feedback. Good luck and good sending.

- Spencer Kollas
Director of Delivery Services at StrongMail Systems

more » | comment »

How Opens Are Tracked and Reported

Written on March 12th, 2009 | Posted by admin in Metrics

The eec blog post introducing the new “render rate” (by the Measurement Accuracy Roundtable) has drawn dozens and dozens of comments to date – from supportive to some that question the value of the standardization initiative.

There were also a number of comments and questions that indicate many people still don’t understand what the open rate does and doesn’t measure and how open rates are actually tracked. This blog post will be the first of a series from various members of the Measurement Accuracy Roundtable that address the comments and questions posted on the blog.

Before diving into a detailed explanation of how the open rate works and what it does and doesn’t measure, I wanted to remind everyone of the core objective and purpose of this initiative.

The Measurement Accuracy Roundtable was formed with two primary purposes:
1) To ensure that email industry metrics that were widely adopted accurately measured what they were designed to measure;
2) That the metric was measured consistently by vendors and marketers. The intent was not to eliminate metrics or pose our opinion or preferences on email marketers.

With that background and reminder, let’s dive into the basics of the open rate, which hopefully conveys why the eec took up the initiative to standardize this popular email metric…

How open rates are measured: Your email technology automatically inserts html code that references an invisible (often referred to as a “clear” or “1×1″) tracking image in your email, usually at the bottom of the email.

Like the other images in your HTML emails, they are actually hosted on a server, not embedded within the email. When a recipient opens the email, and images are not blocked, the image is called/pulled into the html message from the hosting server. As the image is pulled into the message, it is appended with a unique identifier that is associated to the receiving email address. That rendering of an image associated to an email address has been commonly referred to as an “open.” Now, it gets more complicated.

When an “open” is counted: With the above definition in place, let’s look at the scenarios in which an open is counted or reported:

  • Images are not blocked when the recipient fully “launches” or opens the email.
  • Images are not blocked when the recipient views the email in a preview pane (a feature on an increasing number of email clients and services).
  • A recipient scrolls through the inbox slowly enough to allow the tracking image to load, even though the email was not actually viewed in full or in the preview pane.
  • A recipient clicks on a link in a text email. This particular email service provider or software counts the clicked link as an open. Even though there is no way to track whether the text message was opened (because it has no tracking image), we assume the recipient had to open the message (or view in preview pane) to view the message or click the link. Note: In this example the email tracking software would report one and open and one click.
  • A recipient clicks on a link in an HTML email that is fully opened, but images are blocked or disabled. The text-email logic applies here.
  • A recipient clicks on a link in an HTML email viewed in a preview page, but images are blocked or disabled. The text-email logic applies here, too.
  • A recipient views an HTML email on an iPhone, iTouch or other mobile device that automatically renders HTML emails with images enabled.
  • A recipient clicks on a link on a text or HTML email on a mobile device that does not render images. The text-email logic applies here.

    When an “open” is NOT counted: OK, with me so far? Now, it gets even more confusing. Here are the scenarios when an open is NOT counted or reported:

  • Images are blocked when the recipient fully “launches” or opens the email.
  • Images are blocked when the recipient views the email in a preview pane (a feature on an increasing number of email clients and services).
  • A recipient scrolls through the inbox so fast that the tracking image doesn’t have time to load.
  • A recipient clicks on a link in a text email. This particular email service provider or software does NOT count the clicked link as an open. In this case the rationale is that although an open can be inferred, it was not actually captured. Thus, the metric is kept “pure” and the open not counted.
  • A recipient clicks on a link in an HTML email that is fully opened, but images are blocked or disabled. The same text-email logic from the previous example applies here.
  • A recipient clicks on a link in an HTML email that is viewed in a preview page, but images are blocked or disabled. Again, the text-email logic applies.
  • A recipient clicks on a link on a text or HTML email on a mobile device that does not render images. The text-email logic applies here, too; thus, no open is tracked. The same text-email logic applies.
  • The HTML or text version is read on a BlackBerry or similar mobile device that does not render images.
  • An HTML email is viewed on a Blackberry (as above) and is later opened in Gmail (or other email service/client) with images blocked. The email has been opened and read twice, but no open has been counted.

    I could probably come up with more scenarios that show how inconsistently an open is or isn’t counted or reported, but you should have the gist by this point.

    My fellow Measurement Accuracy Roundtable members will contribute a follow-on series of posts to further explain our rationale for the proposed render rate.

    In the meantime, if anyone still doesn’t understand how opens are tracked and reported, please post your question in the comments, and I’ll give it another shot.

    Lastly, I’d like to personally, and on behalf of the entire Measurement Accuracy Roundtable, thank everyone for their feedback and comments posted on the eec blog. Are you really passionate about this and other email measurement topics? Join the eec and our Roundtable!

    - Loren McDonald, Silverpop
    Co-Chair of the Measurement Accuracy Roundtable

    more » | comment »

  • Two-Click Survey Results: What is the primary metric by which you measure the success of your subject line A/B tests?

    Written on July 15th, 2008 | Posted by admin in Metrics, Subject lines, Two-click survey results

    The answer…
    42% –> Open rate
    13% –> Click rate
    18% –> Click-to-open rate
    26% –> Conversion rate

    Are you surprised by the results? Share your comments below.

    Also, visit the eec homepage to answer the latest Two-Click Survey question:
    Are you in compliance with the new CAN-SPAM rules that went into effect this month?

    –>See more Two-Click Survey Results.

    more » | comment »

    'I Was Told There Would Be No Math.'

    Written on May 7th, 2008 | Posted by admin in Metrics

    I am no longer satisfied looking at program dashboards that roll up multiple messages and present average/mean performance numbers without a measure of variance. This wasn’t always the case.

    When I first cut my teeth as an email marketer in the late 1990s, it had been a while since I’d taken a math class. Having attended a small liberal arts school in the Midwest, I was never asked to crunch a single number, and happily concentrated on the subjects that most interested me at the time—Cultural Anthropology and Sociology. So when I went to report on those first campaigns, I didn’t have a whole lot of tools in my belt to analyze them. I calculated the basic performance metrics—open rate, click through, etc.—and when trends or anomalies presented themselves I took notice.

    Since those good old days, I have brushed up considerably on my analytical skills. Now I still rely on many of the same KPIs, and a few more, to judge program performance, but I definitely see the data in different ways.

    As mentioned above, now I can’t look at a set of data without calculating their standard deviation. It’s of course useful to know, for example, what the average open rate is for a group of messages, but without a measure of variance, the average doesn’t yield much useful information. For example, I could have two different campaigns of three messages each that both have an average open rate of 30%. If we stop there, we could deduce that both programs are performing similarly with regards to opens—when one program could be comprised of messages with open rates of {29%, 30%, 31%} and the other could be {3%, 19%, 68%}. The standard deviation of the open rates for the first campaign is 1—not much variance—but the standard deviation of the second is 33.87, indicating some substantial fluctuations. Again, both have mean open rates of 30%, but there are probably different things going on in data set #2 that warrant further exploration.

    If you don’t currently calculate the standard deviation of your campaign metrics, I suggest you try it. Don’t concern yourself that the standard deviation is the “root mean square deviation of values from their arithmetic mean,” just type =STDEV into excel and select the data. It only takes a second, and I believe you may find the information useful.

    —Nicholas Einstein of Datran Media

    more » | comment »

    Platinum Sponsors
    Gold Sponsors
    • Strikeiron-225
    • Logo_harlandclarkedigitalsm
    • Adestra-logo_100x31_
    • Clickmail_marteking_logo
    • Epsilon
    Silver Sponsors
    • Logo_listrak_-_new_2010_-_smaller
    • Logo_leadspend
    • Logo_silverpop_-_new_2009_-_smaller
    • O_eloqua_140x26
    • Streamsendlogo_180x150
    • Acx-di-logo_final_color_small
    • Bwm_logo_eec
    • Compendium-logo
    • Liveintent
    • Logo_getresponse2012