fbpx
Marketing Research

Why marketing experiments are useful in analyzing email marketing?

How are you measuring and analyzing your email marketing communications? Email marketing, like most digital communications, is a great source of information. Depending on your email marketing platform you can find many things to measure and analyze from each email or campaign you send. You can look at metrics like open rate, click rate, conversion rate, etc. But all these metrics have almost no significance if they are not measured against something else.

I have learned a lot Marketing Experiments with email marketing. A few years ago I was lucky to meet a customer analytics company that taught me about control groups and measuring lifts. After discovering this I never saw email communications the same. Having these two concepts always in mind became the way I crafted email marketing strategies from that day on.

So, what do marketing experiments have to do with this? Using a control group is a way to do experimental design. The name for this type of research design is “Before-After with Control Group”. In these types of tests, you can measure the effect that the manipulations have on your dependent variables (sales, conversion rate, visits to stores, etc.), and you can compare these results against what happened to the control group, and the difference between the two is called the lift.

Let’s say you want to measure the average ticket of customers who receive a certain email. If you would send your regular email to your targeted segment, without creating a control group, you would have the result in a few days. Let’s say the average ticket for those customers was $25. Is this good or bad? The answer is you don’t really know because you don’t have anything to compare it to.

So here is where it gets interesting!

Once you have your customer segment or sample that has the characteristics to receive the email communication, you now can calculate from that segment a random control group and take them out. Your control group should not receive the email, or any type of communication during the time you will be conducting the experiment.

Once is time to measure the results on the average ticket you will now have the average ticket of the customers that DID receive the email and the ones that DIDN’T. The difference between those two numbers will be the LIFT the email had on the average ticket of your customers.

There are many things you can start experimenting on with control groups within marketing. This was an example of something that has been very useful to me and has made my team and I have a better view of the results of one of the many channels we have in marketing. I hope this was useful!

Author Image
andreaizas

Leave a Reply

Your email address will not be published. Required fields are marked *