Segmentation analysis is just a fancy way of saying “don’t generalize.”
Many email marketers pour over open rates, click-through rates, and conversion rates daily, hoping for that subtle yet undeniable improvement, fretting over any drop in engagement or lack of improvement.
Unfortunately, these same marketers are most likely measuring everyone’s engagement from a single mailing and comparing it to everyone’s engagement from the previous mailing, or the previous weekly, monthly, or quarterly average.
This method, while useful for a quick glimpse of performance, gives the marketer only a vague understanding of what actually drives engagement and conversion.
Did the users who didn’t click on this email not like the offer? Was it not big enough? Was the email unappealing? Was it sent at the wrong time of day?
All of these things can affect whether or not a user clicks or converts. More often than not, however, there are more impactful factors involved. Does the recipient trust the brand? Is she in market for the product? Did she purchase the item being promoted or another item that serves the same purpose recently? Each of these factors could cause a recipient to simply ignore the message, no matter how great the offer or awesome the creative.
So how can a marketer know that the increase in performance on this message is due to the offer or creative instead of the recipients’ current needs?
One answer is by conducting segmentation analysis.
Segmentation Analysis — The Basics
Imagine a crowded Peet’s Coffee (way better than Starbucks, in my opinion). You’re trying to sell some women’s yoga pants. You have a good supply, enough for everyone in the store.
You shout, “Hey, I’ve got yoga pants! Who wants some? $39.99 each, normally $45. Act now!”
Let’s say there are 1,000 people in the store, and 20 people decide to buy some pants. Sweet! Fifty others asked about the pants, but decided not to buy.
So, you saw a 7 percent interest rate (similar to click-through rate) and a 2 percent conversion rate. Cool beans.
But imagine what it would be if you measured differently. Imagine if you added a simple level of segmentation to your analysis.
When you look at your sales more closely, you see that 18 females and two males decided to buy. Turns out there were 500 females and 500 males in Peet’s at the time (this is a mega Peet’s, and a popular one at that). So now you discover that, although your conversion rate for the entire population is 2 percent, your conversion rate in the female segment is 3.6 percent (18/500) and your conversion rate in the male segment is 0.4 percent (2/500). Obviously, the female segment performed much better.
Imagine now that you look further into the female group. You’re going to use a psychographic segmentation this time. You’re going to ask all the women to identify themselves as sedentary, active, or extremely active. Then you’re going to see who purchased based on those groups
Here’s what it looks like when you’re done:
|Extremely Active||80||7||8.75 percent|
So it turns out that extremely active women were very responsive to your offer and your pants. Cool! Not to be forgotten are the active women. What are these two groups looking for? Is it different? Finally, is there something the sedentary women were looking for that they didn’t see? What about the men?
Ultimately, the objective with segmentation analysis is to get a better understanding of who responded, and how the differences among these groups can help you better market to them in the future.
You must avoid analysis paralysis, of course. You can slice the data 100 different ways, but you only need to go as far as will help you test a treatment to drive better results. Bottom line: Conducting segmentation analysis in your email campaigns can help you understand how and where to focus your marketing efforts to achieve maximum campaign performance.
Email usage is on the rise, shifting to mobile to create an “always on” email culture. How does email marketing change? The ... read more
Email marketing can be overwhelming, especially if you are new. Lots of tweaking, designing and testing. These eight tools will help you ... read more