Glassdoor is built to protect employee speech, and it defends that position hard. Reviews still come down when they breach the community guidelines. Reviews that are simply unflattering do not, and a company that presses too hard on those usually ends up with a story about pressing.
Profanity aimed at a person, discriminatory or harassing language, threats, confidential business information, and content naming individual non-executive employees. Glassdoor prohibits naming most staff outright, which makes this the single most successful flag category we file.
Competitors, applicants who were rejected, contractors misrepresenting their status, and reviews that describe a role, office, or system your company does not have. Glassdoor will look at these when the flag comes with a coherent factual account, and internal records help establish it.
"Management is incompetent" is opinion and stays. "The company withheld March wages" is a factual claim that payroll records can disprove, and a flag citing that specific falsity has a route. Framing matters more here than on any other review platform — general complaints about a review get closed without a second look.
One departed employee posting repeatedly under different accounts, or a coordinated set posted in a single week after a restructuring. Glassdoor enforces one review per person per year per employer, and the pattern is usually visible in timing and phrasing.
A truthful, negative account of what it was like to work at your company stays, and should. Where the pattern in your reviews reflects something real, removal work is the wrong spend. We will say so, and point you at the response and employer-brand side of review management instead.
Send the employer profile. Within 48 hours you get each negative review classified against the specific guideline it breaches, or marked as protected opinion with the reason. Includes an honest read on whether the review pattern is a content problem or a workplace one.
We name the target reviews, the guideline cited for each, the internal records we need from HR to support the flag, and the reviews we are deliberately leaving alone. You sign before anything is filed.
Flags go in one at a time with a specific guideline reference, never as a bulk complaint about tone. Where a first pass is rejected we re-file against a different clause once, then stop. Weekly diff: removed, rejected, pending, current rating.
Close-out report with the outcome of every flag. Where the rating needs rebuilding, the route is a compliant internal campaign — asking all employees, not selected ones, with no incentives. Related work is described in our case studies.
No, and we do not try. Glassdoor defends reviewer anonymity in court and unmasking attempts are expensive, usually unsuccessful, and reputationally worse than the review. Any firm claiming it can identify a reviewer is describing something you should not buy.
You can invite employees to share their experience, provided you ask everyone, do not incentivise it, do not script it, and do not review drafts. Anything narrower breaches Glassdoor’s guidelines and tends to be visible in the review pattern.
Clear guideline breaches, particularly reviews naming individual employees, often resolve within one to two weeks. Contested cases run four to eight weeks. Glassdoor’s decisions are final in practice after a single re-review.
Naming senior leadership is permitted, as executives are treated as public-facing. The flag then depends on the content — false statements of fact, harassment, or disclosure of personal information rather than the naming itself. Send us the review and we will tell you which applies.
A senior analyst reads the profile and returns a guideline-by-guideline verdict, including the reviews we would not touch.
A genuine customer complaint stays on Google as well; what comes down is the review that breaks a written rule.
Reddit keeps most of what is written about a company, so the realistic target is often the ranking rather than the thread.
Trustpilot removes what fails verification and keeps the rest, including reviews the business is certain are unfair.
A critical video stays up unless a privacy, copyright, or harassment ground actually applies to it.
A published article rarely disappears; a correction and a de-indexed URL are the outcomes worth pursuing.
A conviction that stands stays visible, but the copies scattered across mugshot sites can be reduced to the official record.
The same framework we use in week-one of every engagement — how to classify your branded SERP, spot removable content, and prioritize what to fix first. Free PDF, no sales call.