The statistic that makes your testimonial worse

There’s a result in the giving research that stops people the first time they meet it. One group was shown a photograph and a few sentences: Rokia, a seven-year-old in Mali, is going hungry, and your donation will feed her. A second group got facts instead — millions of people across the region face food shortages, this many children malnourished, this much of the harvest lost. The first group gave roughly twice as much. That much you might have guessed. The part you can’t guess is what happened to a third group, who were shown both — Rokia’s story and the statistics together. They gave less than the people who saw only Rokia. The numbers didn’t reinforce the story. They drained it.

This is the identifiable-victim effect. One vivid individual moves us more than a faceless many, and — the uncomfortable twist — the many can move us less than the one, even though the many is objectively the larger tragedy. Paul Slovic, who has spent a career on this, is fond of a line attributed to Mother Teresa: if I look at the mass, I will never act; if I look at the one, I will. Compassion doesn’t scale with the size of the problem. Past a single face, it flattens out, and sometimes it bends back down.

⁂

Bring that to the page where you’re trying to convince someone your product is worth their trust. Most of us reach instinctively for scale: teams save 30% of their week, trusted by 12,000 companies, a tidy row of logos. Scale has its place — but it’s pulling a different lever than you think. “Trusted by thousands” is safety-in-numbers, the social-proof reflex: everyone else is in the water, so it’s probably fine. That’s a real force and a different essay. The identifiable-victim effect is almost its opposite. It says the single, specific, named person — with a situation you can picture and a before-and-after you can follow — will out-persuade the aggregate, and can be quietly weakened by sitting next to it.

So the practical shape of a case study changes. Open on one real person doing one real thing. Not “teams recover hours of admin time” but one named customer, the specific mess they were in, the specific afternoon they got back. A name, a situation, a concrete change. Let the reader run the arithmetic that matters most, the one the research keeps finding: that person is a lot like me. And be wary of the move almost every landing page makes — bolting a big anonymous number directly onto the most human story on the page. The study is a warning that the number can cool the story down rather than heat it up.

⁂

Here’s where it breaks, because it does. Aggregates are not worthless; they carry a different kind of weight. A B2B buyer doing real diligence wants the number — the retention figure, the customer count, the uptime — and will distrust a pitch that’s all anecdote and no evidence. The effect is strongest on the first, emotional read, and weaker once someone shifts into deliberate evaluation. Add to that an honest limit: this literature grew up in charitable giving and response to harm, not SaaS testimonials, so treat the carry-over as directional rather than settled. The lesson isn’t delete your numbers. It’s don’t lead with them, and don’t glue them to your best face. Give the statistic its own room, a beat later, for the part of the reader who has started counting.

A wall of logos tells a visitor that a lot of people made this choice. One person, named, stuck and then unstuck, tells them something the brain has always found easier to believe — that the same thing could happen to them.