The Hidden Cost of a Bad Glassdoor or Indeed Review on Hiring
The Hidden Cost of a Single Bad Glassdoor or Indeed Review on Recruiting
Last updated: June 2026
A single bad Glassdoor or Indeed review rarely looks like a crisis on the day it posts. There is no press call, no regulator letter, no inbound from the board. It is just a star rating and a few hundred words from an anonymous account.
Then a recruiter notices the applicant pipeline thinning. Time-to-fill creeps up. A finalist passes on the offer. A second finalist asks for more money than the band allows. The cost shows up everywhere except in the place that caused it.
This piece breaks down what one bad review actually costs in 2026, why that number is larger than most HR leaders assume, and what an employer can and cannot do about it.
What candidates actually do before they apply
Job seekers in 2026 do not behave like job seekers in 2016. They research first and apply second.
Glassdoor’s own data, reported across recruiting industry coverage, shows that roughly 86 percent of candidates read company reviews and ratings before deciding whether to apply, and that users read about six reviews before forming an opinion. Independent research from Software Advice puts the number in a similar range, with most job seekers describing reviews as central to their decision. Indeed reports that visitors to a company page spend more than a quarter of their time inside the reviews tab.
The harder numbers concern what happens when those reviews are bad.
A CareerArc study found that only one in five candidates would consider applying to a company with a one-star rating, and that women were 33 percent less likely than men to apply to a one-star employer. A widely cited Inc. analysis of that same data set put offer-rejection at 69 percent of job seekers when employer reputation is poor.
In other words, a bad review is not a vanity problem. It is a top-of-funnel problem. Candidates who never reach the application stage do not show up in any ATS report, and they are the ones the data says are walking away first.
The numbers behind a single bad review
The cost of one review is hard to see because it does not have its own line item. It hides inside three numbers HR teams already track.
1. Cost per hire. SHRM’s 2025 Benchmarking Report, as covered by industry press, puts the average U.S. cost per hire at about $5,475 for non-executive roles and roughly $35,879 for executive roles. A drop in qualified inbound applicants forces the same role to be sourced through paid channels, agencies, or referral incentives instead of organic apply-clicks, and every one of those routes is more expensive than a candidate who arrives through the front door.
2. Time to fill. SHRM’s 2025 data puts median U.S. time-to-fill near 44 days, with senior roles regularly exceeding 90 days, as summarized in SHRM’s own commentary on staffing metrics. Industry analyses cited alongside that report estimate vacancy costs of $4,000 to $9,000 per month per open role in lost productivity, overtime, and project slippage. Every week a bad review keeps a candidate from clicking apply is a week the role stays open.
3. The reputation premium. This is the number most leaders have not internalized. A Harvard Business Review and ICM Unlimited study covered in LinkedIn Talent Blog’s research summary found that a company with a damaged reputation has to pay roughly 10 percent more in salary to convince a candidate to take the job, and that only about 28 percent of candidates would take the offer even at that premium. Scaled across a 10,000-person company with normal turnover, that premium can reach the millions of dollars per year in extra compensation that buys nothing except a willingness to overlook the reviews.
Multiply any of those by the number of roles a given negative review touches over its indexed lifetime, and the cost of a single review stops looking like a comms issue.
Why one review is rarely one review
The other reason the cost is bigger than it looks is that a single review does not stay in one place.
Glassdoor and Indeed both rank well for “[company name] reviews” queries on Google. Many candidates also type the company name directly into Google and skim the first page, where the same review can appear as a snippet, a knowledge-panel pull, or an autocomplete suggestion. Backlinko’s CTR research shows that the first organic result captures roughly 28 percent of clicks, with rapid decay after that, which means a bad review parked in the top three slots is being read by a meaningful share of every person who searches the company name for any reason.
AI search has made this worse. Reddit, Glassdoor, Indeed, and similar user-generated platforms have become primary sources for answer engines. Search Engine Land’s citation studies and reporting from the Columbia Journalism Review document how often these platforms are quoted directly in AI Overviews, ChatGPT, Perplexity, and Gemini answers. A candidate who asks an AI assistant “what is it like to work at [company]” is now likely to receive an answer that paraphrases the worst review on the file.
So when an executive asks “how much can one Glassdoor review really hurt us,” the honest answer is that it is rarely one Glassdoor review. It is one review feeding a Google SERP, an Indeed page, a Reddit thread, and an AI answer simultaneously, for as long as it stays live.
Why “just respond to it” is not a strategy
Most employer-relations playbooks tell HR to respond professionally and move on. That is good hygiene, and Glassdoor’s own research does show that the share of candidates whose view of a company improves after seeing a thoughtful response is meaningful.
It is also not enough on its own, for three reasons.
First, the response sits underneath the review. The candidate still reads the review first, and a response cannot unsee a sentence like “the director screamed at me in front of the team.” The CareerArc and CareerArc-adjacent data on offer rejection holds even when responses are present.
Second, the FTC’s 2024 Consumer Reviews and Testimonials Rule now constrains what an employer can do to influence the review file at all. The final rule, in effect since October 2024, prohibits soliciting reviews from employees or their relatives without conspicuous disclosure of the relationship, prohibits insider reviews that fail to disclose the connection to the business, and authorizes civil penalties of up to roughly $53,088 per violation. The old playbook of asking the team to post a few positives to dilute a bad one is now a federal compliance problem.
Third, the legal route to remove a defamatory review is slow and uncertain. Glassdoor and Indeed are protected from liability for user content by Section 230 of the Communications Decency Act. Glassdoor publishes a policy of resisting subpoenas seeking reviewer identities, and reports successfully defending anonymity in over 100 cases. Removal generally requires either a community-guidelines violation under the platform’s own terms, or a court order after a “John Doe” defamation suit, as outlined by employment counsel like Vorys. That is a real route. It is not a fast route.
What an actual playbook looks like
Three things have to happen in parallel.
1. Stabilize the platform pages. Respond to the review on the platform with a short, calm, specific reply that does not relitigate the facts. Flag any review that violates the platform’s content policy, including reviews that reveal protected information, identify the reviewer in ways that breach privacy, or contain provable factual claims that can be disproved with documentation. Use the platform’s flagging flow; do not threaten the reviewer publicly. Where a review crosses into provable defamation, engage counsel and document the chain that supports a takedown request before sending one.
2. Suppress the SERP and AI citations. For most employers, the realistic objective is not to delete the review. It is to push the URL hosting the review off the first page of the brand-name search and out of the citation set that AI assistants reach for. That is a content-and-authority discipline: publishing employer-owned pages, executive thought leadership, structured profiles, and third-party coverage that out-rank the review for the brand query, and using AI search reputation work to influence which sources answer engines treat as authoritative. Justin’s team treats AI search reputation as a separate workstream from traditional SEO suppression because the citation logic is different. Our AI Search Reputation Management and Suppress Negative Search Results 2026 pages walk through the mechanics.
3. Fix the upstream signal, not just the downstream review. Most companies with one bad review have several. CareerArc’s reporting that nearly 38 percent of fired or laid-off employees post negative reviews of former employers tells you the input function is predictable: bad separations produce bad reviews. Employer-side fixes to the exit process, the management-training program, and the candidate-experience loop reduce the rate at which new bad reviews enter the system. Suppression without that upstream fix is treadmill work.
For a company actively losing finalists to a single review or to a cluster of them, the response sequence usually combines Content Removal work on any defamatory or policy-violating items, Business Reputation Management work to rebuild the SERP and AI citation footprint, and a Company Crisis Management engagement if the underlying issue has begun bleeding into press or social.
What we guarantee
Digital Crisis Management works only on outcomes that can be measured. For employer-reputation cases, that means specific page-one rankings, specific AI assistant answers, and specific platform-level removals where the policy facts support them. We put those outcomes in writing, with timelines, and we do not bill the full engagement on cases we do not believe we can resolve.
If you are watching a bad review drag a role open week after week, or watching offer-acceptance soften across an entire function, the cheapest day to start is the day you notice it. Every additional week the review sits at the top of the brand-name SERP and inside the AI citation set is another week the company is paying the 10 percent reputation premium on every offer.
Talk to us when you are ready. The consultation is free and the conversation stays inside our walls.



