How to Optimise Your Careers Page for AI Search

Your careers page was built for humans using Google. AI doesn't work the same way.

When ChatGPT, Perplexity, or Google AI Overviews answer questions about your company, they don't "visit" your careers page the way a candidate does. They parse machine-readable data, scrape text content, and synthesise answers from multiple sources. If your careers page is a Workable widget with no additional content, AI has almost nothing to work with.

This guide covers every technical step to make your careers page AI-visible — from quick wins you can implement today to the full optimisation stack.

---

Step 1: Check if AI can even access your site (5 minutes)

Before anything else, check your robots.txt file at yourcompany.com/robots.txt.

Look for these user agents being blocked:

User-agent: GPTBot
Disallow: /

User-agent: ClaudeBot
Disallow: /

User-agent: Google-Extended
Disallow: /

User-agent: PerplexityBot
Disallow: /

43% of UK employers block at least one AI crawler. If you're blocking them, nothing else in this guide matters until you fix this.

The fix: Either remove the AI-specific blocks entirely, or selectively allow access to your careers content:

User-agent: GPTBot
Allow: /careers
Allow: /jobs
Allow: /about
Allow: /blog
Disallow: /admin
Disallow: /internal

User-agent: ClaudeBot
Allow: /careers
Allow: /jobs
Allow: /about
Allow: /blog
Disallow: /admin
Disallow: /internal

This gives AI access to the content that benefits you while keeping internal pages private.

---

Step 2: Create your llms.txt file (1-2 hours)

An llms.txt file at your domain root tells AI models who you are as an employer. It's the single highest-impact action you can take.

Place it at: yourcompany.com/llms.txt

Template:

# [Company Name]

> [One-line description — what you do, how many people, where]

## About
[2-3 sentences: industry, size, headquarters, founded year, mission]

## Working Here
[Specific culture description. Avoid generic statements like "great team."
Instead: "Engineering teams work in 2-week sprints with full autonomy over
technical decisions. Most people are in-office Tuesday-Thursday, remote
Monday and Friday."]

## Compensation
[Department-level salary ranges. Be specific.]
- Engineering: £[range]
- Product: £[range]
- Sales: £[range]
- Operations: £[range]

## Benefits
- [Pension contribution: X%]
- [Annual leave: X days + bank holidays]
- [Health: private medical / dental / etc.]
- [Remote policy: X days remote]
- [Learning budget: £X per year]
- [Other standout benefits]

## Career Progression
[How people grow. Promotion cadence, levels, lateral moves.]

## Tech Stack (if relevant)
[Languages, frameworks, infrastructure]

## Open Roles
See current vacancies: [URL]

## Links
- Careers: [URL]
- LinkedIn: [URL]
- Blog: [URL]

Tips:

• Be specific. "Competitive salary" is useless to AI. "£65K-£85K for senior engineers in London" is useful.
• Update it quarterly. Stale data is worse than no data.
• Keep it under 2,000 words. AI doesn't need a novel.

---

Step 3: Add JSON-LD structured data (half day, engineering)

Structured data is the machine-readable markup that AI models trust most. Add these schemas to your careers page:

Organization schema (on your careers landing page)

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Your Company",
  "url": "https://yourcompany.com",
  "description": "What your company does",
  "numberOfEmployees": {
    "@type": "QuantitativeValue",
    "value": 250
  },
  "address": {
    "@type": "PostalAddress",
    "addressLocality": "London",
    "addressCountry": "GB"
  },
  "sameAs": [
    "https://linkedin.com/company/yourcompany",
    "https://twitter.com/yourcompany"
  ]
}

JobPosting schema (on each job listing)

{
  "@context": "https://schema.org",
  "@type": "JobPosting",
  "title": "Senior Software Engineer",
  "description": "Role description...",
  "datePosted": "2026-03-15",
  "employmentType": "FULL_TIME",
  "jobLocationType": "TELECOMMUTE",
  "applicantLocationRequirements": {
    "@type": "Country",
    "name": "UK"
  },
  "baseSalary": {
    "@type": "MonetaryAmount",
    "currency": "GBP",
    "value": {
      "@type": "QuantitativeValue",
      "minValue": 75000,
      "maxValue": 95000,
      "unitText": "YEAR"
    }
  },
  "hiringOrganization": {
    "@type": "Organization",
    "name": "Your Company",
    "sameAs": "https://yourcompany.com"
  }
}

The salary field is the most important. Companies with structured salary data have AI salary accuracy of 84%, versus 22% without.

---

Step 4: Make your careers page content-rich (ongoing)

The biggest mistake: treating your careers page as a job listing aggregator. A page that's just a Workable/Lever/Greenhouse embed gives AI nothing to cite except the raw job specs.

What AI needs on your careers page:

1. A culture section with specific, concrete details (not "we're passionate about innovation")
2. Employee count and locations — basic facts AI can verify
3. Benefits summary — the full list, not just "competitive package"
4. Working arrangements — remote/hybrid/office, core hours, flexibility
5. Career progression — how people grow, example career paths
6. Team descriptions — what each department actually does

Content format matters:

• Use heading hierarchy (H1, H2, H3) — AI parses document structure
• Use lists for benefits and requirements — more parseable than paragraphs
• Include specific numbers — "250 employees across London and Manchester" not "a growing team"
• Keep content in HTML, not loaded via JavaScript — many AI crawlers can't execute JS

---

Step 5: Publish salary ranges (policy decision)

This is the single most impactful action for AI accuracy, but it requires a business decision.

The case for transparency:

• AI salary estimates are wrong 78% of the time for companies that don't publish
• Wrong (usually low) estimates cause candidates to self-select out
• UK pay transparency legislation is heading this direction anyway
• Companies with published ranges see 30% more applications (LinkedIn data)

If full transparency isn't feasible:

• Publish ranges on your careers page even if not on individual listings
• Use bands: "Engineering roles range from £50K-£120K depending on level"
• At minimum, include ranges in your llms.txt file

---

Step 6: Create supporting content (ongoing)

Blog posts and articles on your company site become citation sources for AI. The highest-value content for AI visibility:

• "What it's like to work at [Company]" — directly answers the most common AI query
• "[Company] Engineering/Product/Sales Team" — team-specific pages that AI can reference
• "[Company] Benefits Guide 2026" — comprehensive, annually updated
• "[Company] Interview Process" — candidates ask AI this constantly

Each piece of content should be:

• On your domain (not Medium, not LinkedIn)
• Crawlable (not behind a login)
• Specific (not generic employer brand messaging)
• Updated (AI deprioritises obviously stale content)

---

Step 7: Monitor your AI presence (weekly)

After implementing changes, measure the impact:

1. Ask AI models your own candidate queries — "What's it like to work at [Company]?", "What does [Company] pay for [role]?"
2. Compare responses to reality — Are salaries accurate? Is the culture description current? Are benefits correct?
3. Check cross-platform consistency — Do ChatGPT, Perplexity, and Google AI tell the same story?
4. Track changes over time — AI models retrain regularly. Your presence will shift.

Or use a monitoring tool (like OpenRole) to automate this across all platforms.

---

The implementation timeline

Expected result: A company scoring 34/100 on AI visibility can reach 70+ within 4-6 weeks by following this guide.

---

See where you stand

Run a free AI employer brand audit and get your score across all major AI platforms. See exactly what candidates hear when they ask AI about your company.

30 seconds. No signup.

→ Run your free AI employer brand audit

---

Sources: OpenRole audit data (500 UK employers, 2026), SparkToro (2024), Pew Research (2025), LinkedIn Talent Solutions salary transparency data (2025), Schema.org documentation.