In 2020 someone visited the New Bedford Whaling Museum, an hour south of Boston, and left a one-star Google review with the words, worst aquarium ever. The museum is not an aquarium. It is a museum about the history of whaling, with the skeletons and the scrimshaw and the Melville connections you would expect, and at the time it contained no live marine animals whatsoever.
Six years later, that review is a T-shirt. It is also a sweatshirt, a crewneck and a tote bag, printed with a one-star rating and a minimalist whale skeleton, and the product description commits fully to the joke. The shirts have sold out.
Doubling down on the story, the museum’s chief engagement officer, Annelise Conway, delivered the memorable line, “We do not have live whales here, so we are actually the worst aquarium ever.”
It is a delightful story, and it has been picked up everywhere from local Boston television to national outlets. It is also about to be copied badly by a great many organisations, because the lesson most of them will take from it is misguided.
How to respond to a bad review
Making a joke at your own expense is riskier than it looks, and there is research that sets out how the risk works. Maurice Schweitzer, a Professor of Management at the Wharton School, with Brad Bitterly, then a Wharton researcher and Alison Wood Brooks, a Professor at Harvard Business School, ran eight experiments on what humour does to status. Their finding looks at when it works, and when it backfires..
Attempting humour always signals confidence, whether the joke lands or not. What decides whether your standing rises or falls is the second signal, competence. Humour that works signals both at once and lifts status. Humour that misfires signals confidence paired with incompetence, and that particular combination does real damage.
Now apply it to a whaling museum printing its own worst review on a T-shirt. The confidence signal is automatic, because nobody amplifies a criticism of themselves by accident. Everything therefore rests on what the move says about competence, and this is where the museum was on safe ground.
The complaint was not that the staff were rude, or the building cold, or the ticket overpriced. The complaint was that a whaling museum had failed to be an aquarium, which is not a flaw in the museum at all. Repeating it costs the institution nothing, because no reasonable visitor is deterred by the news that a history museum is a history museum.
So the shirt lands as confidence and competence together, which is the combination that Wharton and HBS professors say raises status. It says: we know exactly what we are, we know you misunderstood, and we are secure enough to enjoy it. The joke is the message.
When responding to a negative review with humour backfires
Now consider the opposite case. In 2009, Domino’s Pizza was running dead last in consumer taste tests. The criticism everywhere, and very direct. Focus groups said the crust tasted like cardboard and the sauce tasted like ketchup. Worse, as the incoming chief executive Patrick Doyle later described, blind testing showed consumers liked the pizza less when they knew it was from Domino’s.
Domino’s did not print “tastes like cardboard” on a T-shirt. It would have been a catastrophe if they had, and the Wharton framework says exactly why. The joke would have signalled plenty of confidence, but it would have paired that confidence with an admission of incompetence, since the criticism was true. Turning an accurate complaint into a punchline tells customers you find their disappointment amusing.
Instead the company aired the criticism in its own advertising, put the focus group footage on television, and reformulated the recipe. Doyle’s framing at the time was that spinning things no longer works. “Great brands going forward are going to have a level of honesty and transparency that hasn’t been seen before.”
Fourth-quarter profits more than doubled as customers came back to try the new pizza.
Both organisations did something braver than the corporate default, which is to argue with the reviewer, flood a comments section with positive reviews, or when possible delete the comment. But they did opposite, and the variable that decided which was right is not personality or tone of voice. It is whether the complaint was true.
A simple rule for handling bad online reviews
There’s a diagnostic anyone can apply the next time a bad review lands. Ask two questions in order.
Is the criticism accurate? If yes, the only move is Domino’s. Acknowledge it plainly, say what you are changing, and change it. Humour here reads as contempt, and customers can tell the difference between an organisation laughing at itself and one laughing at them.
If the criticism is not accurate, ask the second question: is it about something central to what you actually offer? A review complaining that a whaling museum lacks live whales is peripheral, in the sense that it describes a mismatch between the visitor’s expectations and the institution’s purpose rather than a failure to deliver that purpose.
That is the space where a joke signals competence rather than undermining it, and where a critic can be turned into a marketing department. A false accusation about something central, a claim of poor safety or bad faith, belongs in neither category. It needs a correction, not a punchline.
The implication is that the New Bedford strategy is available far less often than the coverage suggests. Most bad reviews are at least partly true, and most organisations reaching for a witty comeback are reaching for it because they would rather not fix the underlying problem.
Do online reviews reflect real quality?
Even for the reviews that are wrong, the numbers matter more than they should. At the University of Colorado Boulder’s Leeds School of Business, Bart de Langhe, Philip Fernbach and Donald Lichtenstein examined 1,272 products across 120 categories and found that average user ratings correspond poorly with objective quality, showing little convergence with Consumer Reports scores, resting frequently on sample sizes too small to be informative, and failing to predict resale prices. Ratings ran higher for expensive products and premium brands regardless of measured quality. Consumers, meanwhile, lean on the average star rating far more heavily than on any other cue.
A star rating can be dragged down by a visitor who came to the wrong building, and the people reading it will treat the number as a verdict on quality.
A useful finding comes from Georgios Zervas at Boston University Questrom School of Business, working with Davide Proserpio then completing his PhD at Boston University. They compared hotels across TripAdvisor, where managers routinely reply to reviews, and Expedia, where they almost never do, and found that hotels which began responding saw their ratings rise. It was not that responses charmed existing critics into changing their minds. It was that once a business was visibly paying attention, customers with a poor experience became less likely to post a negative review in the first place. Responding does not repair reputation after the fact so much as change what gets written.
What a bad review can teach you about your customers
Which brings us back to New Bedford, and to the detail that most of the coverage has swum past. The museum has since added a small aquarium. It was installed years after the review, and the museum now says, with what one hopes is a straight face, that it would like a good rating for that one.
So the reviewer was not simply wrong. They were early. Somebody arrived at a whaling museum expecting to see living creatures, discovered they had misunderstood what the place was, and said so in the bluntest available terms. That is a data point about a gap between what an institution offers and what visitors turn up expecting, and the museum has answered it twice: once with a small tank, and once with a T-shirt that explains the misunderstanding better than any defensive reply ever could.
The T-shirt does the work of correcting the expectation while getting the joke in first. That is the trick worth stealing, and it is a subtler one than laughing off criticism.
Interested in this topic? You might also like…