Take Time to Test

In the tactical digital environment of instant marketing, the temptation is very real to bypass strategic principles of testing first.

It’s not unusual for me to see four variations of a logo on Linkedin, where the host of the post asks their friends which one they like. While seeking marketplace feedback before launching any marketing content is a wise approach, this growing trend falls short. My fear is this practice gives the marketing manager a false sense of adequate testing.

Case in point – how accurate were the polls in the last presidential election? They ranged from predicting a landslide victory in the Electoral College for Hillary Clinton, plus the House changing its majority, to the actuality of a very different outcome that shocked the nation and the world. So how did this happen? Pollsters and political scientists will be arguing the results for decades to come. But what we just learned (or rather, were reminded of) is that people will say one thing but then do another. So should marketers rely on a process similar to polling in an attempt to select the most effective visual tools? They’re doing it, which is bad news. The good news is there’s a better way that is attainable for most everyone.

---

"Case in point – how accurate were the polls in the last presidential election?"

---

With the online platforms available for marketing, the ability to test variations of ads, logos, or any other form of marketing communications exists. Let’s take the example of a new apartment complex opening in a metro market of around 1 million. Instead of running a popularity contest for a pretty graphic, consider conducting a measurable, real-world test of your materials.

STEP ONE

Select four test markets as close as possible to the one in wich you will be marketing. Keep all factors in mind including population, climate, culture, socioeconomics and location within each market. Once you select a test market for each of your variations, set up a post on your Facebook or Instagram page (depending on your demographic) featuring ONE logo for each test. Make sure and use the same copy and call to action for each logo variation so you maintain an “apples to apples” comparison.

STEP TWO

Sponsor each post with your test logo, setting up the same demographics for each, but with each variation geotargeted to a unique location (test market). Assure you click the options to target people who are actually located in the geographic area, rather than those whose accounts are registered in the area. If you are wondering if you should optimize for clicks or impressions, there is no perfect answer. I prefer clicks, as that is the ultimate outcome marketers will be looking for. Plus, clicks from a “learn more” button are more objective to use as a benchmark for measuring results.

STEP THREE

You’re probably wondering at this point, how do I gracefully market a product that does not exist in the area? Well, you can’t. I recommend you set up a landing page for each sponsored post that simply states “Thank you for participating in our market test. We are currently assessing opportunities in your area for future consideration.”

STEP FOUR

After a week, step back and analyze the results. If you do not have a clear winner, you can at least start the process of eliminating the lowest performers. You may need to repeat the test with your top two finalists. If so, I recommend testing them again head to head in a fifth market not originally tested in.

FINAL THOUGHTS

This process will require a few hundred dollars and a week to 10 days out of your schedule to complete. Keep it in perspective; this is a very small investment in time and money compared to the bigger picture of you or your client’s overall expense in product development and launch. Plus, you can move forward with the confidence of knowing your marketing content has been field tested in the real world and proven in front of targeted personas.