TESTING YOUR EMAIL MARKETING CAMPAIGN
It is common in email marketing to conduct split tests to compare the effectiveness of subject lines, creative approaches, offers, etc. Marketers will often use that data to email the rest of the campaign or the next campaign with the version that had either the highest open rate or click-through rate (or both). The reality, however, is that the email opened or clicked-on most may not be the version that produces the best results.
For example, in some ecommerce split tests we've conducted, the email with a lower click-through rate (CTR) will produce a higher number of transactions. The reason frequently is that the lower CTR email was of greater interest to fewer people, resulting on a higher number of actual orders.
|
|
So how do you make sure your stats aren't lying to you? Track results beyond open and click-through rates, all the way through to your desired action -- whether it is a purchase, downloading of a white paper or registering for an event. One of the simplest methods is to use Web site tracking, which entails putting a single line of code on each Web page that you'd like to track. This enables you to track those actions actually taken on your site, such as a visits to an order confirmation page.
Industry average open and click-through rates are XX percent and XX percent. Various companies report "industry average" email metrics such as open, click-through and bounce rates. We all love these statistics, but let's be clear: they are not industry averages. Instead, the stats usually just provide a snapshot into the average results from one company's client base. Unless your profile is similar, the "average statistics" are probably not a good benchmark for your company's program. Factors that influence these "average" statistics?
Types of emails sent (newsletter, notification, ecommerce, etc.), industries, sophistication of senders, size of lists, source of addresses, amount of personalization, nature of relationship with recipients, and how they define their metrics.
What's the optimum way to use these reported industry statistics? As goals rather than as clear benchmarks. If, for example, your newsletter generates a 25 percent open rate, but you learn of a 40 percent "industry average", establish internal targets and action steps to increase your open rate to 30 percent, then 35 percent and beyond.
Open rates are tracked using a transparent one pixel GIF image hosted on a server, just like the viewable images in an HTML email. As a result, any action on the recipient's part that attempts to load the "open GIF" will count as an open. This includes the Preview Pane in Outlook, for example. On the other end of the scale, text messages that are opened but not sent in HTML/multipart format and/or opened by a non-HTML compatible email client will not be reported as an open. There are a host of other reasons an email that is opened may not be reported as such, including images being disabled and the email being opened offline after download.