• by Loren McDonald , Op-Ed Contributor, September 27, 2006

Here's what I wish I could say to the next person who brags to me about the open or click-through rate (CTR) his latest e-mail campaign or newsletter just produced: "Would you just get over the open rate already?"

You're not going to get a raise or a fatter budget because your open rate hit 40 percent. Your CEO wants the top and bottom line: How much revenue do you bring in? Are you earning back what the company spent on you? But if your e-mail marketing program isn't meeting these and other business objectives, open and click-through rates certainly won't tell you what's wrong.

Instead, you need to run some diagnostics on your program and determine the key drivers of your program's successes and failures. There are lots of these diagnostic-type metrics, but let's take a look at four examples: share of e-mail, list activity, subscriber drop-off and deliverability. They aren't designed to tell you if your e-mail program met its goals. But they will help you uncover problems or weaknesses and develop strategies to improve them.

Most diagnostic metrics use standard message stats like opens, clicks, conversions, bounces and unsubscribes, but it is important to move beyond the individual message and plot say, six months or 10 messages, on a graph and study them over time to spot trends up or down.

Take share of e-mail. Sure, you got a 15 percent CTR on your last message, and maybe you even know that's up 5 percentage points over your last six messages. But do you know which links (products, offers and articles) consistently drove the most traffic or conversions?

Maybe you have 20 total links in your last message. If you focus just on the CTR itself, you might miss a key indicator, like finding out just two of those links drove 75 percent of your activity. Knowing that can help you redesign your e-mails to highlight the links (content and offers) that matter most to your customers and subscribers.

Look at list activity, too. Do you know how many readers are still opening and clicking six months after they sign up and which ones are basically just valid addresses in your database?

When we've analyzed inactive-subscriber reports for clients, we see lists that are typically between 33 percent and 75 percent "dead." One financial-services client recently found out 58 percent of its main list hadn't opened or clicked on a single e-mail in the last six months! It was quite a staggering metric to them. But once you know that, you can deploy various means to re-engage your long-term subscribers.

You also need to know your subscriber-interest timeline, which shows you when interest drops off to the point where subscribers don't actively open e-mails or act on them. Typically we see this dropoff after the first few messages or months, particularly if a program is not "front loaded" during the welcome process and initial messages.

You can actually measure when subscribers start to lose interest by comparing activity according to opt-in date. Once you know when drop-off typically begins, you can develop incentives or strategies to keep subscribers engaged longer.

Finally, deliverability is still a challenge for most e-mailers. It has never been harder to get messages both delivered to the inbox and rendered correctly. You first need to check your delivery rate over several months or messages, as reported by your e-mail service provider, delivery-monitoring service or list-management software.

Is delivery going down steadily or does it ebb and flow? You'll need to dig into more data to find out what's going on, such as unsubscribe, bounce and spam complaint rates at each ISP.

Yes, even though these are just a few examples of statistics, it is still a lot of data to sift through. But if you're not looking at metrics like these, frankly you are probably pretty clueless as to what's actually going on with your e-mail marketing program.

OK, I feel better now. But please, don't ask me what the average e-mail open rate is, or you might see steam rising out of my thinly covered head.