Email marketers today are very concerned about their “open” rates, but what exactly is an “opened” email and how should marketers interpret the open rate?
The “open” metric sounds like a very concrete measurement, but in reality it is not really “black and white.” The reason that there is some ambiguity is because there is no sure-fire way to tell if a recipient actually opened an email, much less determine what they did if they did open it.
Almost all email marketing tools measure an “open” by embedding a tiny invisible image somewhere in the body of an HTML email. The image is usually hosted on the same server that sent the email. When the recipient opens the email, his email program makes a request to the server to retrieve the image file so that it can be displayed along with the rest of the email. When the image is requested, the sending server records that a request was made for the image, and reports that the email was “opened” in whatever metrics reporting tool that is available to the sender.
Sounds like a great system, so what could possibly go wrong? As it turns out, there are several things that can interfere with accurate “open” metrics reporting. First of all, almost all email clients block the display of images by default. This was implemented years ago as a response to the proliferation of pornographic spam before spam-filtering was as robust as it is today. The idea is that readers may not want to display images until they have a chance to see what the email is about and who sent it. One of the biggest problems with using images to record “opens” is that many email programs are very subtle about telling the reader that images are being blocked, and unless there are other large images in the email that are also blocked and display the familiar red “x” of a unloaded image, users may not ever see any reason to unblock images in your email. For that reason, its a good idea to include several images in your email that will cause the email to render noticeably different without images so that the reader is more likely to unblock images, thus allowing your “open” metric to be more accurate.
Another issue with using HTML images to determine “opens” is that there is no way to determine *who* actually opened the email. If the user forwards the email to several people and they all open the email (and do not block images), each open will be recorded as an “open” in your metrics and thus could skew your interpretation. The same thing happens when the same user opens the same email several times. Each open is recorded. Many metrics tools also keep track of a statistic called “unique opens” to try and filter out opens by the same recipient. The problem with this is that the metric tool attempts to filter these by identifying the IP address of the opener, which can be the same for many different readers if they are all behind the same firewall. This is usually not a problem, as each image is unique, but is recipients forward the email to others in their organization and they open the email also, their opens will not contribute to the “unique opens” metrics despite the fact that they are indeed “unique.”
In addition to these issues, there is really no way to determine an opened email for “text only” emails. You may not think that this applies to you if you only send pretty HTML emails, but you would be mistaken. Almost all email marketing tools send both an HTML and a text-only version of every email because they do not know what the capabilities of the receiving email program might be. The email program then determines which version to display based on the capabilities of the program and the preferences of the reader. Some recipients (though admittedly, not many) will set their email program to display text-only for all senders or unknown senders, or because they like the increased speed and security of text-only email (explaining how that works is a whole other email). Also, many mobile devices are only capable of displaying text-only emails, though this is rapidly changing with newer smart phones like iPhones and Blackberries. In any event, as an email marketer, you should note that some small percentage of recipients will have text-only readers and you will never be able to determine if they open your email (unless they respond to your call to action).
Because of the inability to determine if a reader truly opened an email or not, it is possible for your metrics tool to show that a specific user clicked on a link (or converted in some other way), but didn’t open the email. This is, of course not possible, but there is no way around this illogical way to report the metrics, so don’t call your email marketing provider and complain when you see this type of metric.
In the end, the “opened” metric is helpful for helping gauge the effectiveness of your email marketing efforts, but you cannot rely on that metric alone. It is only truly helpful when viewed in conjunction with other metrics like clicks and/or conversions. Making sure that you pay close attention to details like these does make email marketing ore challenging, but the results are worth usually it.
