• by Stephanie Miller , May 5, 2009

In my panel on email measurement this morning, we talked about how taking an "engagement methodology" approach is simple in concept, but difficult in execution. Dela Quist, CEO of Alchemy Worx, walked through analysis he did on his own email newsletter. Taking open and click data and trending it over seven issues gives some pretty interesting insights. For example, while the issue open rate is about 50%, the open reach over time is closer to 80%. So if all you have is the issue open rate, to try to improve it you might test subject lines or day of week - pretty broad, gross tactics. However, if you know that only 20% of your audience was not opening over time, then you might focus on those specific subscribers and test content, subject line, cadence or even frequency. It's a much finer tool for taking action.

What I love about Dela's example is that it's done with just open and click data. Data that is readily available to every marketer, although not typically as an automated trend report. Similarly, at Return Path we see that trend data of complaint data (complaints are measured everytime a subscrciber clicks the Report Spam button) by message type, day of week, domain and subsccriber vintage can reveal some very interesting actions that are not apparent if you just look at the average complaint rate.

Ted Wham of Orbitz asked a very good question at this point. "What ESP can give me this report, automatically, and without extra steps?" The answer is very few, although both Strongmail and Axciom raised their hands to indicate that they have these types of reports available.

A lively discussion ensued, where Dela challenged ESPs to provide this level of reporting. And also challenged marketers to ask for reports of this nature. "You will get it if you keep asking for it!" Of course, at the end of the day, it's a shared responsibility between the marketer and the vendor.

The other panelist, Allen Dickson of Overstock.com, talked aboot how they pull this data manually and produce their own custom reports which help ensure that measurements are tied to business goaals. He feels that doing this manually gives him the power to pull in data from other systems, increasing the relevancy of the analysis.

Why do this analysis at all, especially if your current "batch and blast" program works well enough? Simply, if you don't have data to back up your arguments, then it's very difficult to get out from under the daily pressure to promote the daily deal with blasts to the full file.

For our own clients, we find at Return Path that doing this engagement level analysis even just once a quarter can make a material difference in our ability to advocate for tighter subscriber experiences - which is the only way to improve response and consistently reach the inbox.