Email Experiments: Put Yourself in Your Contact Lists

There’s just no substitute for first-hand experience. Nowhere else is this more profoundly true than in the science of email marketing, especially when carrying out tests and experiments with things in your campaigns. Metrics and stats reveal a lot about what you’re testing for, but there are some things which analytics can’t easily measure or predict. In some testing situations, you need to become a recipient in your email marketing contact lists and actually see for yourself.

There are some areas in your email campaign that metrics can’t give the complete picture of. For example, if you’re split testing for the optimal placement of calls-to-action for various email clients, numbers like open rates and click-through rates can only reveal your recipients’ actions which (although very reliable) are merely indirect ways to measure what you want to find out. Instead of guessing how your emails really appeared, why not do the next best thing and send the test emails to yourself?

Besides testing for email client compatibility, other tests like evaluating design factors, checking for validity of links, identifying ISP specifics, analyzing segmentation strategies, etc. require gaining first-hand insights to fully grasp the underlying issues. So the next time you’re testing for these and similar campaign components, try adding your email address to your targeted mailing list and applying the five tips below to get the most out of your test:

1. Use a test email address.

Before you start including your email address into your test lists, make sure that it’s a separate address that you’re not currently using in other professional situations. Aside from being unique, your test email address should also be easy to spot for list cleanup. In addition, you need to have multiple test addresses for multi-segment testing.

2. Be part of every segment.

A basic A/B split test configuration is comprised of at least three segments from a given list. Of course, the numbers can exceed this figure, depending on the properties being tested. The main point here is, regardless of how many segments the test uses, it’s vital that your own email addresses appear in all the segments (i.e., one unique address per segment).

3. Know what you’re testing for.

This seems to be commonsensical enough to be mentioned here – and it is. But it’s far from being in common practice. More often than not, email marketers tend to fall for the cause-and-effect mix-up and mistakenly test for the result instead of the driver. Understand what you’re trying to determine and walk through the process before your run your tests.

4. Keep track of your actions.

Once the test is underway, you need to precisely record and continuously monitor the actions that you perform on your test emails after you’ve received them. Take note of opens, clicks, scroll, etc. that you do in relation to the emails in question. You’ll see why in the next point.

5. Adjust your test analytics.

This is where all the recording and monitoring you did in the previous point become useful. As a test participant, your actions contribute to how your test analytics and metrics were calculated, which means the initial test results you obtain aren’t completely reliable. Using your recorded actions from point #4, you have to make specific adjustments to your metrics to essentially remove your participation in the test (at least on paper).

The next time you run your tests, go beyond the numbers and try seeing things in your email campaign for yourself. After all, seeing is believing.