Imagine living on an island with 5,000 people, only one of whom spoke your native language. That person would probably become a fast friend, right?

Now imagine instead of an island of 5,000, you share a native language with only one other person on earth. That’s one out of seven billion.

Surely then you’d want to chat occasionally.

Or not. In Ayapa, Mexico, live the last two native speakers of a language called Ayapaneco. They live half a mile from each other, but haven’t spoken to each other in years… and have no desire to.

Their story surprised me at first, until I thought about it a little bit. Just because they can speak to each other doesn’t mean they have anything to say to each other. Just because the outside world views them primarily as the last two remaining speakers of Ayapaneco doesn’t mean they see themselves that way.

This story was particularly interesting to me because of my job: I’m a market researcher. I spend a lot of my time trying to find patterns in data that I can translate into insights, which my clients use to communicate effectively with their audiences.

Oftentimes, this takes the form of segmentation, or looking at different subsets of data to find commonalities and differences. This can range from basic demographic comparisons (young vs. old, male vs. female, etc.) to sophisticated modeling that includes demographics, attitudes, and behaviors.

But here’s the thing… Anytime we build a profile, we’re basically assuming that the people in that profile act similarly, that they have other things in common beyond what we’re using to bucket them. And that’s where it can get dicey, as our two speakers of Ayapaneco demonstrate.

What if we business owners are making assumptions based on the wrong data points?

This isn’t a purely academic argument. Think about Millennials, for instance. We make huge assumptions about someone’s behaviors, attitudes, and lifestyle based on nothing more than birth year. And you can’t go a week without seeing some new article about the quirks and foibles of some generational group, followed the next week by an article suggesting that maybe we’re not so different after all.

By the way, the Pew Research Center has done a good job of looking at generational differences, and even put together a nifty little quiz that will tell you how “Millennial” you are. I surprised myself with a score halfway between Gen X and Millennial – much higher than I thought I would be.

None of this is intended to knock generational segments, or demographic segments, or any kind of segments for that matter. Anything you can do to effectively target your customers is going to help you be more efficient and effective. However, anything you do to target customers also means you’re assuming certain commonalities within your target profiles.

And that’s the point. If you’re going to assume, assume responsibly. What does that mean? Constantly challenge and test your assumptions. Are you creating a service aimed at “high-income singles”? Don’t assume you can slap a label on your target customer and be done with it. Dig deeper and try to find out the real motivations for interest in your offerings. What is it about being high-income that makes them interested? What is it about being single that makes them interested? Why is the combination of high-income and single so powerful? Who else shares those needs?

Look for the real similarities hiding under the superficial ones.

Remember, good marketing starts with a solid understanding of your prospects – in their terms, not yours.  The more descriptive you can be in profiling prospects, the more effective you can be in communicating with them.