Social proof vs herding

Social proof is the tendency to look to others when deciding what to do. There’s nothing manipulative about that in itself, and people have decided this way for millennia. You’re in a new city, you see a restaurant full of locals, and you infer it’s good. That’s social proof working as it should.

In CRO it shows up in a few shapes: testimonials from real customers, case studies with specific numbers, “X people bought this today” counters, review stars and counts. The ethical version uses real signals that people can check and that help them decide.

Herding is social proof without the proof, the look of consensus where there isn’t any. The most common examples are fake scarcity signals: “only 2 left in stock” on a product with 200 units, or “50 people are viewing this” on a page that’s had 50 visitors all week. These do work in the short term, because they trigger the same circuits as real scarcity, but they wear away trust the moment someone spots them. And users are increasingly wise to them.

The alternative is real numbers from real events. If 12 people bought a product today, say 12. If it was zero, say zero. If stock is low, say so and name the variant that’s running low. Being specific is what makes you believable, and that trust is what converts over the long term. Good social proof would still be useful even if it made your product look less popular.

Faking it trades trust for a single transaction. The cost never shows in the current session, only later, when the customer doesn’t come back.

Put it to work

Audit your social proof. For each claim, check it is true, verifiable, and relevant to the decision in front of the user.

Your answer

Show the worked example

There is no right answer. Write yours first, then compare.

Sources

1. Cialdini - Influence: The Psychology of Persuasion
2. Baymard Institute - Social Proof in Ecommerce