How We Improved Review UX By Not Asking People To Pretend
An honest first impression matters to us — even when it's rough.
May 11, 2026
All major review platforms ban profanity in reviews.
The problem — users have to correct themselves, run their words through tools, and sanitize what they actually felt before a platform will even accept it.
“Please avoid using profanity or attempts to approximate profanity with creative spelling, in any language. Comments and media that include hate speech, discriminatory remarks, threats, sexually explicit remarks, violence, or the promotion of illegal activity are not permitted.”
“Keep it clean! We request that contributors refrain from using vulgar or profane terms. Of course, we don't allow the 'super profane' words and our general rule of thumb is, if you wouldn't shout it in public, don't say it here.”
“There are a number of reasons why Tripadvisor rejects or removes reviews, ranging from community standards violations (such as the use of profanity) to fake review activity.”
“Content that is violent, graphic, threatening, demeaning, insulting or harassing.”
Each platform tells users the same thing: your experience must be translated into corporate-friendly language before anyone hears it.
We do not encourage profanity. However, we value the truth — and we built everything possible to extract it from any text.
How it works
A user submits a contribution in any form — any language, any emotional intensity, any tone. Our AI processes the input and extracts structured signals: pest_infestation, host_unresponsive, deposit_refund_dispute. Each signal has a confidence score and contributes to the venue's aggregate Risk score.
The original text never appears publicly. It stays in our system as evidence — accessible only to the venue owner (for disputes) and to the contributor themselves. Other users see only the structured output: scores, signal patterns, aggregated data.
Privacy and anonymity
We do not store personally identifiable information. Contributions remain anonymous — you will never be associated with a review you wrote in a moment of frustration.
The single exception is a court order. If a court requires disclosure, we provide the full text. But because we don't store PII (personally identifiable information), linking a specific review to a specific person is structurally not something we can do.
Why this matters
It has never been more important to speak openly — sometimes without filters, without the lens of “constructive feedback.” Self-censorship protects the industry from criticism, not the user from harm.
When reviews are tied to public identity, people filter. When they filter, signals are lost. When signals are lost, the next traveler walks into the same trap.
We don't ask people to become a better version of themselves to leave a review. We accept them as they are — and do the work for them.
Goler is trust infrastructure for hospitality. Learn more about our scoring methodology, product policy, and dispute process.