Marketleap - An Acxiom Digital Resource
Login Services Contact About University SEO Tools Home
Marketleap University
Search Engine Opt. 101
the Marketleap Report Blog
the Marketleap Report
Sign-up
SEM Resources
Free Search Engine Marketing Tools
Link Popularity Check
Search Engine Saturation
Keyword Verification

 


The Marketleap Report
Volume II, Issue #14 - Part 2, August 21st, 2002

The Hunt for Stats
Tracking Email and Clear Pixels
By Keith Boswell

People don't like to be hunted. You never want to feel that someone has tagged your ear like an elk and now your migration pattern is available to be traced and studied. It's unsettling.

It brings back Cold War type fears when you learned that satellites could track your every move if they wanted to. Your phone could be bugged by an unscrupulous third-party and they would tape every word, mumble or breath you released into your telephone.

Several of you wrote in inquiring about the capabilities of email tracking after reading the last Report. This is a topic that doesn't get much coverage and in the interest of educating our readers I'll tell you what others aren't. I'm including the responses that drove this issue at the bottom.

Email tracking sounds threatening at first. You wonder, "What are they using the information for?" Before we answer that, let's talk about the type of information we are able to collect from HTML based emails and how it is collected.

When a marketer uses a publishing platform that incorporates campaign performance data, they are accessing a limited amount of information. Graphics that are embedded within HTML messages provide data that can help a marketer determine a campaigns performance.

These graphics (aka: pixels, web bugs, or clear GIFs) provide simple tracking in email or basic HTML. A clear GIF is an invisible graphic that is usually 1 pixel by 1 pixel in size, hence, "I use pixel tracking technology."

When a user opens an HTML based email or HTML web page, graphics (pixels) are requested from the site hosting the HTML. This data includes:

  • The IP address of the computer that is requesting the graphic
  • The time the page containing the graphic was viewed
  • The type of mail client or web browser that received the graphic
  • If a marketer is using cookies, they could also check the identification number of the cookie or update a cookie they had already placed if they choose to

Because email marketers can see if a graphic has been requested and successfully sent to a user, they count that as an "open". Right now the technology is only able to track opens in HTML based emails. If you are using an email application, like Outlook, that provides a preview pane, if any graphic loads it is counted as an open. In text based messages, you can only track the number of clicks that occur on links embedded within an email.

In the case of quickly scanning an email, if a single graphic loads during the time the email is open, it will register as an open. There is no way to determine if a message has been deleted, only if it has been opened. For the marketer to assume that the message was deleted is a bad assumption. Some of our email clients see opens as late as four weeks after the original send date.

Marketleap reports a series of statistics for our readers in an effort to demonstrate the capability of email tracking and to let readers know how they are responding to various issues. The first four numbers that we publish are the number of emails sent, the percentage of emails that bounced back, the percentage of readers who unsubscribed and the number of forwards using the tell a friend option.

Stats for the last Marketleap Report sent July 25, 2002

  • Sent - 884 messages
  • Bounce Rate - 3.28%
  • Unsubscribe Rate - 0.90%
  • Forwards - 1
  • Unique Open Rate - 53.85%
  • Click Rate - 5.66%
  • Unique Click Rate - 5.66%
  • Unique Click/Open Rate - 10.50%

There are then four other statistics that we publish to help gauge the newsletter success.

  • Unique Open Rate refers to the unique number of readers who opened the newsletter. We also track (but don't publish) the number of total opens which means that if someone opens the newsletter more than once, they register as an additional open, but not a unique open.
  • Click Rate refers to the number of clicks generated from the newsletter. This number represents all clicks, including those users who click more than one link.
  • Unique Click Rate refers to the number of clicks without counting duplicate clicks from a single user.
  • Unique Click/Open Rate is a percentage based on the number of unique opens for the newsletter that also produced a unique click for any of the links in the article. This is also considered to be the overall response rate to the message.

The system that Marketleap uses allows us to publish HTML, AOL and Text versions of our newsletter. Depending on what type of software the reader is using, the appropriate email is then delivered to their inbox.

Marketleap's tracking system is built around ad serving technology so we can even update an email on the fly. We can publish updated content for HTML messages at any time during the course of an email campaign. If someone hasn't opened the message yet, or they open it again, they will see the updated content instead of the content that was originally published.

Often times, 3rd-party advertisers on a web site will request clear pixel tracking to ensure that advertising reports, provided by the site selling ads, are consistent with data received from the pixel tracking. In this case, it acts as an audit to ensure accuracy.

In some cases, this type of tracking can be trouble. Spammers can use HTML based emails to see if an email is a valid address by seeing if the message is opened. They can then add that address to a list of live addresses. Some may also use Clear GIFs or invisible graphics to set or read cookies on a recipient's computer. In this case, the marketer is looking at what information the cookie has collected since it was last looked at, such as the sites they have visited.

In most cases tracking is simply used to help marketers gauge the success of their online efforts. Respecting the privacy of the reader is paramount to successfully engaging in this type of tracking. Marketers have to be up front and direct with what they are using tracking information for and what they won't use information for. It's our responsibility to educate while we lead.

If we don't disclose what information we are tracking the backlash will be more dramatic than simply being honest up front. Permission marketing is still the buzz. It's critical that those you market to know exactly what they are giving you permission to do.


Responses to the last Marketleap Report:

Hi,

This newsletter really triggered my curiosity. I hope you will excuse me if this is a little off the mainline topic.

I was wondering about this "tracking" of email. This is a topic that I have never seen covered, and I would be very interested to know up to which extent it can technically be done. What can be seen and how? Is it actually possible to see whether an email was opened, for how long, whether it was disposed of, etc.?

Gee, I didn't even know you could be sure the email even arrived until you get a reply back, or an email confirmation of receipt! (and the last does not event work with all mail clients, to my knowledge)

Yours,
-Sylvain


Hi Keith,

Thank you for your latest report. That was a very accurate description of the (sad) state of customer service right now.

Reading through your article, I was wondering how you (technically) can track if a mail actually was read (or scanned) or deleted right away. Do you have any idea where I could find more information on this?

I would also like to know what certain terms in your Marketleap Report Stats refer to. For example: Unique Open Rate, Click Rate, Unique Click Rate and Unique Click/Open Rate. Can you help me with this? Thank you!

Best Regards,
-Elke

We're interested to hear what you think? Drop us a line if you have any other questions or feedback. We'd also like to know if you're interested in seeing the full range of reporting that we have access to. If so, we'll put up a link to a sample report in the next issue. So friends, til next time, when we're back on the regular gravy train.