The average marketer worth their salt can scratch out their customer’s rough demographics while sitting alone in their room, pecking data points into a spreadsheet.
Creating a profitably converting segment is easy. But just because the segment you lay out in Excel converts profitably, that doesn’t mean your segment accurately defines your core customer base.
Is Your Segment What’s Actually Delivering?
How could that be? If your segmented campaign delivers, then clearly your segment created your marketing success, right?
Well, correlation doesn’t always imply causation.
You see, the qualities you created for your segment aren’t the only qualities certain members of your customers share with each other. You may profitably sell an app to a segment of 25–34 year old males who love Apple products and have already bought 10 apps in the last year, but some of them will also share the fact that they’re single, upper-middle class, and living in major East Coast cities.
Guess what? It’s possible your real customers aren’t actually 25–34 year old avid iPhone users. Rather, your real customers are single, upper-middle class East Coast urbanites.
In this case, your marketing campaign is profitable because enough members of this second demographic fall within the segment you created and not because your segment itself was on-point.
Never assume the segment you create defines your actual customer base. It’s always possible your actual customers just happen to appear in high-enough concentrations within the segment you created to produce profitability.
Too Complicated for Pen and Paper
Believing your segment causes profitability, and doesn’t just correlate with profitability, is a common and easy mistake. Your ideal customers are constructed by data patterns that are a lot more complicated and diverse than anyone could ever think up on their own because us humans tend to fixate on just one or two characteristics we can digest.
By contrast, our top customers use Adobe Analytics to run more than 100 dimensions through the mill when identifying what their profitable segment actually looks like. These dimensions are all evaluated as a whole to determine their relationship to each other and their influence on your desired marketing outcome. And when you look at over 100 variables at once, you’re going to see correlations defined by seemingly unrelated attributes and dimensions, all connected in ways you would never have guessed on your own.
Even if you plug 100+ data points into your spreadsheet, you still wouldn’t be able to find these connections, and you certainly wouldn’t be able to perform next-generation segmentation solutions, including predictive segmentation, with any efficiency. Yes, it’s possible to perform statistical analysis in Excel, but when you use spreadsheets you’re still inputting and calculating your analysis manually, which naturally limits the size, sophistication and creative capabilities of your analysis.
Although a lot of marketing problems can be solved with a little personal ingenuity, accurate segmentation just isn’t one of them. If you encounter problems with your segment, then you must rigorously and continuously test your assumptions about who you’re actually selling to, and you have to use a modern analytics solution to do so.
What’s Wrong with “Good Enough”?
If your segment delivers a profit, then why worry about testing its validity? Do you really need to know why your campaign delivers a profit?
Let’s ignore the fact that accurately knowing your audience will help you create a more profitable marketing campaign. In many ways, the data you pull from testing your marketing assumptions have positive ramifications that ripple out far beyond simply knowing who to shout your product’s praises to.
For example, if you don’t market to your real customers, then there’s a good chance your product development team acts with similar misinformation. When you figure out who your customers actually are, your product development team can then optimize their product to best suit those people who want it the most.
Segmenting is the marketing equivalent of building a product in the lab without testing that product on real users—you may come up with something that sells, but ultimately you’ve only taken one step toward crafting success. And when it comes down to it, a willingness to keep taking additional steps forward often separates wildly successful campaigns from campaigns that are merely “good enough.”