A client sent over their new segmentation debrief last week and the lead persona is named Ashley. Ashley is thirty four. Ashley is a marketing manager who hikes on weekends, listens to podcasts on her commute, values authenticity, and is short on time. There is a stock photo of Ashley holding a coffee.
I have been handed some version of Ashley perhaps forty times.
She is always in her early thirties. She is always slightly time poor and slightly aspirational. She always wants brands to be honest with her. And she always, without fail, resembles the median person sitting around the table that approved the research.
Which is exactly the problem.
A persona deck is not a picture of your customer. It is a picture of your team, wearing your customer's name tag.
This does not happen because researchers are lazy. It happens through a chain of small, reasonable decisions, and I have made every one of them.
It starts with the screener. Someone on the brand side writes the recruiting criteria, and those criteria come out of the assumptions the team already holds about who matters. You cannot discover a customer you did not invite to the interview. If the screener says household income above eighty thousand and shops the category monthly, you have already decided that the occasional buyer and the price sensitive switcher do not exist.
Then comes selection. Thirty interviews produce hundreds of usable quotes and somebody has to choose twelve. The quotes that get chosen are the ones that sound like insight to the person choosing, which means they sound like the way that person already talks.
Then comes the deck itself. Personas get designed to be liked. They go into a room full of executives who need to see themselves as reasonable people serving reasonable people, and a persona with genuinely inconvenient behavior does not survive that room. The customer who buys your product out of habit, or ironically, or because it was the only one left on the shelf, gets edited out because she makes the strategy harder to write.
You cannot discover a customer you did not invite to the interview.
The fix is not more rigor inside the deck. It is one uncomfortable slide. Put the persona next to a plain snapshot of the people who commissioned the study, and make the differences visible. If there are no differences, you did not do research. You did a mirror.
The best consumer insight I got last year came from a Tuesday at karaoke, watching a man in a Blazers jacket sing Total Eclipse of the Heart with absolute commitment while the host called him baby and the whole bar went up. Not one person in that room was Ashley. Every one of them buys things.