Why your review platform profile doesn't matter anymore
For fifteen years, employer review platforms were the definitive source of employer reputation. Candidates checked reviews before applying. HR teams obsessed over star ratings. Entire industries sprouted around "review platform management."
Then AI happened. And traditional review sites became almost irrelevant overnight.
The citation data
Profound analysed 680 million citations across major LLMs between August 2024 and June 2025. The results upend everything we assumed about where AI gets its employer information.
ChatGPT's top sources: Wikipedia (7.8%), Reddit (1.8%), YouTube (1.2%)
Google AI Overviews: Reddit (2.2%), LinkedIn (1.3%), Wikipedia (1.1%)
Perplexity: Reddit (6.6%), YouTube (2.0%), Wikipedia (1.8%)
Traditional review platforms: Not in the top 10 for any major LLM.
That's not a typo. The platforms that defined employer reputation for a generation are barely a footnote in AI's source material.
Why AI ignores review platforms
The reason is technical, not editorial. Legacy employer review sites use robots.txt to explicitly
block all major AI crawlers — GPTBot, ClaudeBot, Google-Extended, and others.
They do this to protect their proprietary review data.
The unintended consequence: AI models can't see these reviews at all. When a candidate asks ChatGPT "What's it like to work at [company]?", the model assembles its answer from whatever it can access — Wikipedia stubs, Reddit threads, random blog posts, and LinkedIn profiles.
Blind and Comparably have similar restrictions. The entire ecosystem of employer review platforms is invisible to AI.
The Reddit effect
Reddit's prominence in AI citations isn't accidental. Reddit signed a $60 million annual deal with Google to license its data for AI training. OpenAI has a similar arrangement. This means unfiltered, anonymous Reddit threads carry more weight in AI responses than your carefully curated careers page.
A disgruntled ex-employee's Reddit rant from 2023 may now be the primary source AI uses to describe your company's culture to candidates in 2026.
What replaces review platform management
If traditional review sites don't matter to AI, what does? The answer is structured, machine-readable data that AI models can directly ingest:
llms.txt
A structured file at your domain root that tells AI models exactly how to describe your organisation. Culture, benefits, salary ranges — in your words. Read our guide →
JSON-LD structured data
Machine-readable schema markup on your careers page. AI models trust this more than any review platform because it comes directly from you.
AI visibility monitoring
Weekly checks on what each AI model actually says about you — and whether it's accurate. You can't manage what you don't measure.
The uncomfortable truth
Most HR teams are still investing time and money in review platform responses, review solicitation campaigns, and employer profile optimisation on platforms that AI cannot read.
Meanwhile, the channel that actually shapes candidate perception — AI — is being fed uncontrolled data from Reddit, Wikipedia, and whatever scraps it can find on your website.
The shift doesn't require abandoning review platforms entirely. They still matter for human-driven research. But the balance of influence has tipped. And the companies that recognise this first will control their narrative while competitors wonder why their application rates are declining.
Sources
- Profound — 680M LLM citation analysis (Aug 2024–Jun 2025)
- Major review platform robots.txt files — verified Feb 2026
- Reddit–Google data licensing deal — $60M/year (Feb 2024)
- OpenAI–Reddit partnership announcement (May 2024)
- Blind, Comparably robots.txt — verified Feb 2026