Bought reviews: what it costs when it comes out

Three failures, not one

The legal exposure is the part everyone names first, and it is real: section 5b(3) UWG treats it as material information whether and how a trader ensures that published reviews come from consumers who actually used or bought the service. A trader who buys reviews cannot answer that question truthfully and cannot avoid it either.

The second failure is arithmetic and the third is pattern. Both are worth more attention than the first, because they explain why the purchase does not even deliver what it promised.

Why the arithmetic defeats it

Start from a realistic problem. An average of 3.9 across 40 reviews is 156 points. Reaching 4.0 needs 4 five star reviews. Reaching 4.5 needs 48, because each additional five star review moves the average by less as the base grows.

So the tempting purchase is not four reviews, it is 48, and 48 arriving inside a few weeks against a normal rate of 3 a month is a twelvefold spike. That is exactly the shape platforms screen for, and it is visible to any customer who sorts by date.

Now add recency. 74 percent of consumers want reviews from the last three months and 44 percent look at the last month. A bought burst fills the window once and empties it, so within a quarter the profile reads worse than before: a high average with nothing recent behind it.

What a purchase promises against what it delivers

Promise	Reality	Source of the problem
A quick jump to 4.5	48 reviews needed from 3.9 at 40 reviews	Arithmetic
Nobody will notice	Twelvefold spike against a normal rate	Pattern
It is a grey area	Disclosure of whether and how is material information	Section 5b(3) UWG
It lasts	Empty three month window within a quarter	74 percent want recent reviews

Arithmetic worked from a 3.9 average across 40 reviews. Consumer figures from the 2026 review survey.

What the same money buys instead

The honest route to 4.0 from 3.9 is 4 reviews, which is under two months at an ordinary rate. That is the campaign worth running, and it clears the threshold that 68 percent of consumers apply.

Spend the rest on the reply rate. 19 percent of consumers expect a same day response, and answers are read by the 82 percent who read AI generated summaries. Consistent replies cost minutes and cannot be bought by a competitor.

Questions and answers

Is buying reviews illegal?

The disclosure duty in section 5b(3) UWG makes whether and how you ensure authenticity material information, and a bought review cannot survive that question honestly.

What about incentivised reviews from real customers?

That is a different case, and it still runs into the disclosure duty. Say what you do and let the reader judge.

How would anyone find out?

The pattern is visible without inside knowledge: a cluster of ratings arriving at once against a known normal rate, and nothing recent afterwards.

What is the honest alternative?

Four five star reviews lift 3.9 to 4.0 at a base of 40. That is under two months at three a month.

Sources

Section 5b Gesetz gegen den unlauteren Wettbewerb (UWG) The disclosure duty for consumer reviews
BrightLocal, Local Consumer Review Survey 2026 Thresholds, recency expectations and response expectations