When criticism helps the review but hurts the reviewer

There is a familiar shortcut for deciding whether a reviewer is trustworthy: look for the person willing to say something bad.

Someone who praises everything has an obvious credibility problem. The critic seems more discriminating, more independent and less interested in keeping a brand happy. A little negativity can look like proof that the praise is earned.

Except audiences may not judge critics that way.

A 2026 paper in the Journal of Consumer Research by Junha Kim and Joseph K. Goodman finds a more complicated split. When people expected reviews to be positive, they tended to see negative reviewers as less credible than positive reviewers. At the same time, they could find the negative review itself more helpful.

The research is not an argument that criticism always costs trust. Its 19 studies, involving 7,330 participants, point to something more conditional: credibility depends partly on whether a message fits what an audience expected from its source. Change that expectation and the effect weakens, disappears or reverses.

The review and the reviewer part company

Study One makes the distinction unusually clear. After preregistered exclusions, 279 university students read either a four-star review praising a vacuum or a two-star review criticising it. The researchers held the vocabulary, length and topic constant, changing the valence of the review. Some participants assessed the review's helpfulness while others assessed the reviewer's credibility.

The negative review scored higher for helpfulness, at 5.86 against 5.51 on seven-point measures. Its author scored lower for credibility, at 4.03 against 4.38.

These are modest differences in one experiment, but they run in opposite directions. The information became more useful while the person providing it lost standing. That is the puzzle at the centre of the paper.

It also explains why a general claim that negative reviews either work or fail is too crude. A review can help someone decide not to buy a vacuum without making them more inclined to trust its author next time. The content has done its job in the moment, yet the source may have less influence beyond it.

For marketers, that separation matters because common indicators can blur the two judgements. A critical post might attract saves, comments or helpful votes because it contains diagnostic information. None of those responses necessarily means the audience has become more confident in the creator. Message value and source value can move apart.

The offence is against expectation

Why would criticism lower credibility when it often looks like evidence of independence? Kim and Goodman use expectancy violation theory to explain the result. People carry socially learned expectations about how others usually communicate. When a message breaks the expected pattern, attention shifts towards the person delivering it: why did they say that and what does it reveal about them?

In many consumer review settings, the expected pattern is positive. The authors' validation work found that ratings for participants' own recent purchases were overwhelmingly high. The claim is not that every audience wants praise. It is that positive evaluations often provide the reference point against which a negative reviewer is judged.

Study Two tested the proposed mechanism with reviews of a hiking trail. Participants rated a negative reviewer as less credible and saw the review as a greater violation of social norms. The statistical relationship through perceived norm violation remained when positive and negative affect were included in the analysis. The authors interpret this as evidence that the response was not simply a bad mood spilling over to the source.

The more decisive evidence comes from changing what people expected. In Study Three, a vacuum from an unfamiliar company entering the market created ambiguity because there was no established consensus about the brand. Under ordinary conditions, the negative reviewer was rated less credible than the positive reviewer. With the new brand, the gap disappeared.

Study Four pushed the test further. Before reading a restaurant review, participants were told that they had heard good things, mixed things or bad things about the business. When prior expectations were positive, the negative reviewer was rated less credible. With mixed expectations, the two reviewers received identical mean credibility scores. When expectations were negative, the pattern reversed: the negative reviewer was rated more credible than the positive one.

The same act of criticism therefore changed meaning with its setting. It could look like a breach of the prevailing norm or a response entirely consistent with it. Under negative expectations, praise became the surprising message and its source paid the credibility penalty.

That boundary condition is the paper's most useful idea. The issue is not negativity by itself. It is the relationship between the judgement, the person making it and the audience's prior beliefs about what a person in that position normally says.

A judgement can outlast the review

Credibility matters because a reviewer rarely speaks only once. Creators, experts and critics build an audience across repeated recommendations. Kim and Goodman therefore tested whether the judgement formed from one review carried into a later decision.

In Study Six A, 420 participants read a positive or negative review of a hiking trail, then encountered a positive review of another trail from the same person. The authors found an indirect relationship: the initial review's valence shaped perceived credibility, which in turn affected willingness to follow the later recommendation. This held even though the second review was positive.

Study Six B moved the outcome to social media. After exclusions, 431 participants saw a positive or negative Instagram post about a hiking trail. Those who saw the negative post rated its author as less credible and reported less willingness to follow the account. A follow-up study removed the credibility questions to check that asking them had not created the result. Willingness to follow was still higher after the positive post, at 4.70 against 2.82.

This was stated intention in an experiment, not observed behaviour on Instagram. The work does not show that one critical post causes a creator to lose followers. It does show that the audience's assessment of a message can alter its relationship with the messenger and that the consequence may surface after the original review.

The wider programme tested the effect across settings including wine, hiking trails, vacuums, kitchen blenders and films. It also varied factors such as reviewer credentials, product complexity and whether people were choosing something or trying to avoid it. One reported replication did not produce a statistically significant effect, which is a useful reminder that the pattern is robust without being universal.

The persona sets the rules

The commercial implication is not that creators should avoid negative opinions. It is that criticism is interpreted through a role.

Consider an established restaurant critic known for exacting judgements and a lifestyle creator known for enthusiastic recommendations. The same harsh assessment may be unremarkable from the first and disorienting from the second. That comparison was not tested directly in the paper, but it follows from the finding that expectations change the credibility effect.

A creator's usual tone is therefore more than a matter of style. Over time, it teaches an audience what sort of message belongs to that person. Category norms matter too. Blunt criticism might be expected in a film review yet feel oddly severe in a recommendation space built around friendly discovery. The audience is judging not only the evidence offered but whether the performance fits the part.

This complicates how marketers assess a potential partner. Reach, engagement and sentiment describe the content or its distribution. They say less about the expectations that sit between a creator and their audience. A useful question is not simply whether the creator is positive or critical, but whether their audience regards that tone as normal for them.

It also offers a caution for personal brands built on contrarianism. Criticism can make an argument sharper and more informative. It can draw attention by breaking a pattern. Yet attention, helpfulness and credibility are different outcomes. A post can succeed on the first two while quietly weakening the third, particularly when the stance feels out of character.

None of this makes consistency an instruction to become predictable. An audience may welcome a change in view when the reasons are clear. The research does suggest that a surprising judgement asks the messenger to carry an extra burden: people begin explaining the departure and some explanations will be about the person rather than the merits of the message.

Credibility lives between message and messenger

We often treat credibility as if it sits inside a piece of communication. Present evidence, acknowledge limits and demonstrate independence, then trust should follow. Those qualities matter, but audiences also compare the message with a social expectation formed before it arrived.

That is why Kim and Goodman's result is more interesting than a contest between positivity and negativity. Criticism can improve the informational value of a review while reducing trust in its author. Under a different expectation, the penalty can vanish or land on the person offering praise.

The critical question is not whether somebody was negative. It is whether that was what the audience expected from them.

What teams need to know

Does the paper find that negative reviews are less credible?

No. It distinguishes the review from the reviewer. In Study One, participants found the negative review more helpful while rating its author as less credible.

When did criticism reduce reviewer credibility?

The pattern appeared when people expected reviews to be positive. It weakened or disappeared when expectations were ambiguous, so negativity alone does not explain the effect.

What happened when people expected a negative review?

The pattern reversed. In Study Four, the negative reviewer was rated more credible than the positive reviewer when participants had first heard bad things about the restaurant.

Should creators avoid criticism?

The research does not support that conclusion. It suggests that audiences interpret criticism in relation to the creator's role, their prior expectations and the surrounding context.

Did the studies measure actual changes in social media following?

No. They measured participants' stated willingness to follow a reviewer or a later recommendation. The paper does not establish real-world follower gains or losses.