How It Works
We inject machine-readable data into your website
so AI cites your facts, not rumours.
When ChatGPT, Perplexity or Google AI researches your company, it searches the web first. Publishing structured data on your own domain maximises the probability AI cites your verified content instead of third-party guesses.
How RAG works (the short version)
Modern AI models like ChatGPT and Perplexity use Retrieval-Augmented Generation (RAG) — they search the web for current information before answering. This means they're not just regurgitating training data; they're actively looking for your content right now.
When a candidate asks "What's the salary at [Your Company]?":
- 1.AI searches the web for pages about your company and salaries
- 2.It ranks results by domain authority and relevance
- 3.It reads the top results and synthesises an answer
- 4.It cites the sources it used (Perplexity does this explicitly)
The problem: Most company careers pages fail this process. They're JavaScript-heavy, lack structured data, and bury key facts in PDFs or behind login walls. So AI defaults to third-party sources — third-party review sites, old Reddit threads — even when those sources are wrong.
What OpenRole does (3 levels)
We inject machine-readable employer data into your existing website. Three levels of implementation — you pick what works for your team. All use your domain authority to increase the likelihood AI cites your verified content.
Level 1: llms.txt
No IT requiredA structured text file at yourcompany.com/llms.txt that tells AI models how to describe your company. Think robots.txt for your reputation. Paste one code block into your site footer.
What it covers: Company overview, salary ranges, benefits, remote policy, interview process, tech stack.
Level 2: JavaScript snippet
Paste 1 script tagA lightweight script (~2KB) that serves structured employer data to AI crawlers visiting your careers page. Makes your company information machine-readable — AI can parse it instantly. Updates propagate without redeploying your site.
What it adds: JSON-LD structured data, dynamic content updates, tracking which AI models visit your site.
Level 3: Your OpenRole profile
AutomaticYour verified employer facts live on your OpenRole profile — a designed, SEO-optimised page that both candidates and AI crawlers read. We control the quality, freshness, and structure. You just keep your data up to date.
What it provides: Canonical source of truth for AI, comprehensive employer facts page, weekly freshness updates, no IT work required.
All levels share the same principle:
Your llms.txt and snippet point AI to your OpenRole profile. When candidates ask about your company, AI finds your verified facts — not third-party guesses. That's what increases the probability of accurate AI answers.
Why this works
RAG-based AI (ChatGPT, Perplexity, Google AI Overviews) searches the web before answering. Publishing structured, machine-readable content on your own domain maximises the probability that AI models cite your verified data instead of third-party sources. Three reasons:
Domain authority matters
When your own domain ranks for employer queries, AI prefers your content over third-party guesses. Publishing on yourcompany.com beats publishing on Medium or LinkedIn.
Changes happen fast
RAG updates in days, not months. One company went from 0% to 11% AI visibility in 2 weeks from publishing structured data. Another went 3.2% → 22.2% with a careers page rewrite.
You already have the data
You know your salary bands, benefits, and interview process. The only problem is it's not published in a machine-readable format on your domain. OpenRole fixes that.
See what AI says about you right now
Run a free audit to see what ChatGPT, Perplexity, Claude and Gemini tell candidates about your company — word for word. Then we show you how to fix it.