Skip to main content
OpenRole
All posts
Guide17 March 2026·11 min read

The Complete Guide to llms.txt for Employer Brands

The Complete Guide to llms.txt for Employer Brands

91% of UK employers don't have an llms.txt file. That means 91% are letting AI invent their employer brand from Reddit threads and Wikipedia stubs.

llms.txt is the single highest-impact action you can take to control what AI tells candidates about your company. This guide covers everything — what it is, why it matters specifically for employer brands, how to write one, and what mistakes to avoid.

What is llms.txt?

An llms.txt file is a plain-text file placed at your domain root (e.g., yourcompany.com/llms.txt) that provides structured information about your organisation for AI models to reference.

Think of it as:

  • robots.txt tells search engines what to crawl
  • llms.txt tells AI models what to say about you

When ChatGPT, Claude, Perplexity, or Google AI encounters your llms.txt, it uses that verified, first-party information as a primary source — instead of guessing from scattered web content.

Why it matters for employer brands specifically

llms.txt was designed as a general-purpose file for any organisation. But it has outsized impact for employer brands because of how candidates use AI:

The top 5 employer queries to AI:

  1. "What does [company] pay for [role]?" (34% of queries)
  2. "What's it like to work at [company]?" (28%)
  3. "What benefits does [company] offer?" (18%)
  4. "Does [company] allow remote work?" (12%)
  5. "[Company A] vs [Company B]" (8%)

Every single one of these can be answered by a well-written llms.txt file. Without one, AI assembles answers from whatever it can find — often Reddit threads, outdated blog posts, or pure fabrication.

The impact is measurable:

  • Companies with llms.txt see AI Visibility Scores 15-20 points higher on average
  • AI salary accuracy jumps from 22% to 60%+ when llms.txt includes compensation data
  • Cultural description accuracy improves significantly when you provide it directly

The employer brand llms.txt template

Here's a comprehensive template designed specifically for employer brands:

# [Company Name]

> [Company Name] is a [industry] company headquartered in [location]
> with [X] employees. [One sentence about what you do.]

## About
Founded in [year]. [2-3 sentences: mission, size, key facts.]
Industries served: [list].
Revenue/funding: [if public or comfortable sharing].

## Working Here

### Culture
[3-4 specific, concrete sentences about your actual culture.
NOT "we're passionate innovators." Instead: "Engineering teams
run two-week sprints with full ownership of their roadmaps.
We do company-wide demos every Friday. The CEO does monthly
AMAs. We're serious about work-life balance — average departure
time is 5:30pm."]

### Working Arrangements
- Office locations: [list with addresses]
- Remote policy: [specific — e.g., "Hybrid: 2 days office (Tue/Thu),
  3 days remote. Fully remote roles available for engineering."]
- Core hours: [if applicable]
- Flexibility: [compressed weeks, flexitime, etc.]

### Team Size & Structure
- Total employees: [number]
- Engineering: [number]
- Product: [number]
- Sales: [number]
- [Other departments as relevant]

## Compensation

### Salary Ranges (GBP, annual, London unless noted)
- Graduate / Junior: £[X]–£[Y]
- Mid-level: £[X]–£[Y]
- Senior: £[X]–£[Y]
- Lead / Principal: £[X]–£[Y]
- Management: £[X]–£[Y]

### Regional Adjustments
- London: base ranges above
- Other UK: [typically X% of London ranges]
- Remote (UK): [policy on location-based pay]

### Additional Compensation
- Bonus: [structure — e.g., "10-20% based on company + individual performance"]
- Equity: [if applicable — e.g., "RSUs vesting over 4 years for senior+"]
- Pension: [employer contribution — e.g., "matched up to 6%"]

## Benefits
- Annual leave: [X days + bank holidays]
- Private medical: [provider/level]
- Dental: [yes/no, details]
- Life insurance: [X times salary]
- Income protection: [details]
- Pension: [contribution details]
- Learning budget: [£X per year per person]
- Wellness: [gym subsidy, mental health support, etc.]
- Parental leave: [enhanced? how much?]
- Cycle to work: [yes/no]
- Season ticket loan: [yes/no]
- [Any distinctive benefits]

## Career Progression
[How people advance. Include:]
- Level framework: [list levels]
- Promotion cadence: [typical timeline]
- Review process: [annual, bi-annual, continuous]
- Lateral moves: [encouraged? supported?]
- Management vs IC track: [both available?]

## Interview Process
1. [Stage 1 — e.g., "30-min recruiter call"]
2. [Stage 2 — e.g., "Technical assessment (take-home, 3 hours)"]
3. [Stage 3 — e.g., "On-site: system design + culture fit (half day)"]
4. [Stage 4 — e.g., "Offer call with hiring manager"]
Timeline: [typical end-to-end, e.g., "2-3 weeks"]

## Tech Stack (if relevant)
- Languages: [list]
- Frameworks: [list]
- Infrastructure: [AWS/GCP/Azure, Kubernetes, etc.]
- Tools: [Jira, Slack, Notion, etc.]

## Open Roles
Current vacancies: [careers URL]
Apply: [ATS URL]

## Links
- Website: [URL]
- Careers: [URL]
- LinkedIn: [URL]
- Blog: [URL]
- Engineering blog: [URL, if separate]

## Contact
Careers team: [email]
General: [email]

Writing tips

Be specific, not aspirational

❌ "We offer competitive compensation and a great benefits package" ✅ "Senior engineers earn £75K-£95K base + 15% bonus. 28 days leave + bank holidays. Private medical via Bupa. 5% matched pension."

Write for machines, not marketers

AI models parse information density, not brand voice. Every sentence should contain at least one verifiable fact. Remove adjectives that don't add information.

❌ "We're a dynamic, fast-paced, innovative team" ✅ "Engineering team of 45, shipping weekly. 12 product squads, each owning a domain."

Include salary data

This is the #1 candidate query to AI and the #1 source of AI inaccuracy. If you include nothing else, include salary ranges.

Update it

A stale llms.txt is worse than no llms.txt. AI will cite your file's data over fresher web sources — so outdated ranges or benefits will persist in AI responses.

Minimum: Update annually. Recommended: Update quarterly, or whenever compensation or benefits change.

Keep it under 2,000 words

AI processes the entire file. Excessive length dilutes the signal. Be comprehensive but concise.

Common mistakes

1. Using marketing language

AI doesn't understand brand voice. It extracts facts. "World-class team building the future of fintech" gives AI zero useful information. "85-person fintech, £12M ARR, London HQ" gives AI everything.

2. Omitting salary data

The single biggest missed opportunity. "Competitive" means nothing to a machine.

3. Forgetting to make it accessible

Your llms.txt must be at the exact domain root — yourcompany.com/llms.txt. Not /about/llms.txt, not careers.yourcompany.com/llms.txt. The root.

Also check that your robots.txt doesn't block access:

User-agent: GPTBot
Allow: /llms.txt

4. Writing it once and forgetting

A llms.txt with 2024 salary ranges actively hurts you in 2026. AI will cite the stale data and candidates will see outdated numbers.

5. Treating it as confidential

Everything in your llms.txt should be information you'd happily put on your careers page. It's public by design — that's the point.

Implementation checklist

  • Write llms.txt using the template above
  • Include salary ranges for all major roles/levels
  • Include specific benefits (not "competitive package")
  • Include current working arrangements (remote/hybrid/office)
  • Save as plain text, UTF-8 encoded
  • Deploy to domain root (yourcompany.com/llms.txt)
  • Verify accessible in browser
  • Check robots.txt allows AI crawler access
  • Run an AI audit to benchmark current state
  • Set calendar reminder to update quarterly
  • Re-audit after 4 weeks to measure impact

Measuring impact

After deploying your llms.txt:

  1. Wait 2-4 weeks for AI models to recrawl and update
  2. Run the same candidate queries you tested before deployment
  3. Compare: Are salary estimates more accurate? Is the culture description closer to reality? Are benefits correctly listed?
  4. Track your AI Visibility Score — expect a 15-20 point improvement from llms.txt alone

Get started

Don't have time to write one from scratch? Run a free OpenRole audit — we'll show you exactly what AI currently says about your company, and where an llms.txt would make the biggest difference.

30 seconds. No signup.

→ Run your free AI employer brand audit

Sources: llmstxt.org specification, OpenRole audit data — 500 UK employers (2026), Profound — LLM citation analysis (2025).