Treat that figure as folklore. It has been copied between blog posts for years, the original context has been lost, and no plausible mechanism explains a near tripling of conversion from any single page element.

The underlying finding holds up anyway, which is the frustrating part. Reviews and testimonials genuinely do change buying behaviour. The evidence is just considerably more modest and more interesting than the number suggests.

What holds up

Testimonials on a landing page improving conversion by something in the region of a third is a figure that reproduces across studies reliably. That is a real effect and worth having. It is not 270 percent.

The more useful findings are about which proof works rather than how much.

Third party reviews outrank testimonials you publish yourself. Visitors discount what a business says about itself and weight what other people say about it, and a review sitting on a platform you do not control carries more weight than the identical words on your own page. Your testimonials still help. They just sit lower in the hierarchy than most businesses assume.

Recency beats volume. A four star review from last month does more work than a five star review from three years ago, because the visitor is trying to establish whether you are still good rather than whether you were once.

Specificity beats enthusiasm. Testimonials containing actual numbers, a named outcome, or a timeframe outperform general praise. Something describing a kitchen finished in nine days against a quoted twelve is evidence. Something calling you professional and reliable is a sentence anybody could have written about anybody.

The perfect rating problem

Products and businesses in the four point two to four point five range frequently convert better than those sitting at a flat five.

This is not counterintuitive once you have been a customer yourself. A perfect score across many reviews reads as filtered, and buyers have learned to be suspicious of it. A handful of mixed reviews among many good ones does what a perfect record cannot, which is establish that the reviews are real.

The practical consequence is that a middling review is not a crisis and suppressing it is worse than displaying it. Respond to it properly and it becomes the most credible thing on the page.

The shortcut is now illegal

The Federal Trade Commission rule banning fake and AI generated reviews and testimonials took effect in October 2024, carrying penalties reported at over fifty one thousand dollars per violation.

Per violation is the phrase worth reading twice. This is not a fine for having fake reviews. It is a fine calculated against each one.

The rule also covers undisclosed paid reviews and testimonials, and reviews written by anybody connected to the business without stating that connection. A friend leaving a glowing review without mentioning the friendship is within scope.

Beyond the legal exposure, fakery is simply fragile. Platforms detect it, customers detect it, and a business caught with invented reviews looks considerably worse than one that had none.

Placement matters more than quantity

The same signal can help on one page and hurt on another, which is the finding most businesses never act on.

Proof works where the doubt lives. On a service page the doubt is whether you can do this specific job, so a testimonial about that specific job belongs there. On a pricing page the doubt is whether this is worth the money, so proof mentioning value or outcome belongs there. On a contact page the doubt is whether anybody will actually reply, so something mentioning responsiveness belongs there.

Scattering the same three quotes across every page satisfies none of those doubts particularly well. Matching each piece of proof to the question being asked at that moment does considerably more with the same material.

What a first year business should do

You have fewer reviews than you want and probably fewer than your competitors. That is a temporary condition and the fix is unglamorous.

Ask everybody, immediately after the work is done and they are pleased. The window is short and it closes within about a week. Most businesses never ask at all, which is why the ones that do accumulate reviews faster than their actual volume would suggest.

Ask for something specific. A request saying it would help if they mentioned what the job was and how long it took produces a usable review. A request asking them to leave a review produces four words.

Get them on a platform first, then quote them on your site with a name and a business where the person is willing. An anonymous testimonial attributed to a satisfied customer is worth close to nothing, because everybody understands you could have written it yourself.

The thing nobody mentions

If you have no reviews and no testimonials, you are not neutral. You are actively worse than a competitor who has six mediocre ones.

A visitor comparing three businesses reads an absence as a signal, and the signal they read is usually that you are new, small, or that nobody was pleased enough to say so. None of those helps.

Which means the first six matter more than the next sixty. The gap between zero and six is where almost all of the value sits, and it is the gap most first year businesses take a year longer to close than they need to.

Be sceptical of the numbers, act on the principle

Treat every conversion statistic you read, including the modest ones here, as directional rather than precise. Most are old, badly sourced, or measuring something other than what the headline claims.

What survives that scepticism is simple enough. People trust other people more than they trust you. They trust recent and specific proof more than old and vague proof. And they notice its absence more than you would like.