AI Employer Brand Report: 500 UK Companies Ranked
AI Employer Brand Report: 500 UK Companies Ranked
This is the most comprehensive study of AI employer visibility ever conducted in the UK.
We audited 500 companies across every major AI platform — ChatGPT, Perplexity, Google AI Overviews, Claude, and Microsoft Copilot. For each company, we asked the questions candidates actually ask: salary, culture, benefits, remote policy, career progression.
Then we scored them.
The headline numbers
| Metric | Value |
|---|---|
| Companies audited | 500 |
| AI models tested | 5 (ChatGPT, Perplexity, Google AI, Claude, Copilot) |
| Questions per company | 6 |
| Total AI responses analysed | 15,000+ |
| Average AI Visibility Score | 34/100 |
| Companies scoring 80+ | 4% |
| Companies scoring below 30 | 46% |
Almost half of UK employers are functionally invisible to AI. When candidates ask about them, AI either says very little or gets it wrong.
The scoring methodology
Each company was scored across five dimensions:
- Discoverability (20 points): Can AI find the company? Does it have llms.txt, structured data, crawlable careers content?
- Salary Accuracy (20 points): How close are AI's salary estimates to actual compensation? Verified against published ranges and industry data.
- Cultural Representation (20 points): Does AI's description match the company's actual culture, or is it outdated/generic?
- Benefits Completeness (20 points): Does AI know about key benefits — pension, leave, remote policy, health, learning budget?
- Cross-Platform Consistency (20 points): Do all AI models tell the same story? Research shows only 25% content overlap between platforms.
Who's leading
The top-scoring companies share common traits:
The top 10% (Score 75+)
These companies invest in machine-readable employer data. They have:
- llms.txt files with comprehensive employer information
- JSON-LD structured data on careers pages and job listings
- Published salary ranges on most or all roles
- Regular careers content that AI can cite
- AI crawlers allowed in robots.txt
Sectors over-represented in the top 10%:
- Technology (particularly companies with strong engineering blogs)
- Financial services (those with transparent compensation)
- Scale-ups (often more agile in adopting new standards)
Common thread: These aren't necessarily the biggest companies or the ones with the biggest employer brand budgets. They're the ones that treat AI visibility as an engineering problem, not a marketing problem.
What the top performers do differently
-
They publish salary data. The single biggest predictor of AI accuracy. Companies with published ranges have an average salary deviation of £3,200. Without published ranges, it's £18,400.
-
They have crawlable careers content. Not just job listings — actual content about culture, benefits, working environment. Blog posts, team pages, "life at" content that AI can index and cite.
-
They use structured data. Organization schema, JobPosting schema, and in some cases EmployerAggregateRating schema. This gives AI verifiable facts rather than forcing it to infer.
-
They don't block AI crawlers. 43% of UK companies block GPTBot, ClaudeBot, or other AI crawlers. Every company in the top 10% allows them.
Who's struggling
The bottom 46% (Score below 30)
Nearly half of UK employers score below 30. These companies are effectively absent from AI — when candidates ask about them, AI either provides minimal information or confabulates.
Common characteristics:
- No llms.txt file
- No structured data on careers pages
- Salary ranges not published
- AI crawlers blocked via robots.txt
- Careers page is a bare ATS widget (Workable, Lever, Greenhouse) with no additional content
Sectors over-represented in the bottom 46%:
- Professional services (outside the Big 4)
- Manufacturing and industrial
- Retail and hospitality
- Public sector (excluding NHS and civil service)
Industry breakdown
| Industry | Avg Score | Top Quartile | Bottom Quartile |
|---|---|---|---|
| Technology | 52 | 78 | 31 |
| Financial services | 44 | 71 | 24 |
| Healthcare (NHS) | 41 | 65 | 28 |
| Professional services | 35 | 62 | 19 |
| Retail | 28 | 48 | 14 |
| Manufacturing | 24 | 42 | 11 |
| Hospitality | 21 | 38 | 9 |
Technology leads because tech companies naturally publish more online — engineering blogs, open-source contributions, transparent career frameworks. They generate the kind of content AI models prefer to cite.
Hospitality and manufacturing lag because their online employer presence tends to be thin — often just job listings on aggregator sites with no first-party content for AI to reference.
The salary accuracy problem
Salary was the most consistently inaccurate dimension across all 500 companies:
- 78% of AI salary estimates were off by more than £5,000
- 62% underestimated actual pay (median understatement: £18,000)
- 16% overestimated
- 22% were within acceptable range (±£5K)
The fix is the simplest one: publish salary ranges. Companies that do see their AI salary accuracy jump from 22% to 84% within weeks.
The reluctance to publish salaries is understandable — internal equity concerns, competitor intelligence, negotiation flexibility. But the cost of AI consistently underquoting your compensation is measurable: candidates self-selecting out before they ever apply.
Cross-platform inconsistency
One of the most surprising findings: AI models rarely agree with each other.
We measured content overlap across platforms — the percentage of information that appeared consistently across ChatGPT, Perplexity, Google AI, Claude, and Copilot for the same company.
Average cross-platform consistency: 25%
This means a candidate asking ChatGPT and Perplexity about the same company will get substantially different answers 75% of the time. Different salary estimates, different cultural descriptions, different benefits listed.
For employers, this means monitoring a single AI platform isn't sufficient. You need visibility across all of them.
What separates a 34 from an 80
The jump from average (34) to leading (80+) doesn't require a massive investment. It requires specific technical actions:
| Action | Typical Score Impact | Effort |
|---|---|---|
| Create llms.txt file | +15–20 points | 2 hours |
| Add JSON-LD structured data | +10–15 points | 1 day (engineering) |
| Publish salary ranges on listings | +10–15 points | Policy decision |
| Unblock AI crawlers | +5–10 points | 15 minutes |
| Publish "life at" content | +5–10 points | Ongoing |
A company scoring 34 can realistically reach 75+ within a month. The biggest gains come from the first three actions — llms.txt, structured data, and salary transparency.
Methodology notes
- Sample: 500 UK employers across 14 industries, weighted towards FTSE 350, major public sector employers, and high-growth scale-ups
- Period: February–March 2026
- AI models: ChatGPT (GPT-4o), Perplexity (with web search), Google AI Overviews, Claude (Sonnet), Microsoft Copilot
- Validation: Salary data verified against published job listings, ONS data, and employer-confirmed ranges where available
- Scoring: Automated scoring framework with manual review for cultural representation
Get your score
Every company in this report has a page on our UK AI Employer Visibility Index. But if your company isn't in the top 500, you can still audit your AI presence for free.
30 seconds. No signup. See exactly what AI tells candidates about you.
→ Run your free AI employer brand audit
Download the full data
The complete dataset — all 500 companies, all scores, industry breakdowns — is available for HR and TA professionals. Contact us for access.
Sources: OpenRole audit data (Feb–Mar 2026), 15,000+ AI responses analysed across 5 platforms. Salary validation: ONS Annual Survey of Hours and Earnings 2025, published job listing data, employer-verified ranges.