The Death of the Careers Page (And What Replaces It)
The Death of the Careers Page (And What Replaces It)
Let's be honest: nobody visits your careers page first anymore.
They ask ChatGPT. They ask Perplexity. They ask Google and get an AI Overview. The answer appears in seconds — salary, culture, benefits, remote policy. No click required. No page visit. No carefully designed hero section with stock photos of diverse teams in glass-walled meeting rooms.
The careers page isn't dead in the sense that you should delete it. But it's dead as a discovery mechanism. It's been demoted from the shop window to the back office.
The numbers
- 60% of Google searches are zero-click (SparkToro, 2024)
- 8% of users click any link when an AI Overview appears (Pew Research, 2025)
- 800M weekly ChatGPT users — a significant share using it for employer research (OpenAI, 2025)
- 80% of job seekers now use AI to research companies before applying (PerceptionX, 2025)
- 1B monthly Meta AI users embedded in WhatsApp and Instagram (Meta, Q1 2025)
The candidate journey has collapsed from a multi-step research process into a single AI interaction. Your careers page is step 4 in a journey that increasingly stops at step 1.
What replaced it
Three things are filling the gap. None of them are particularly sexy. All of them are more important than your careers page redesign.
1. Machine-readable employer data
AI doesn't browse your careers page like a human. It needs structured, machine-readable data:
- llms.txt — a plain text file at your domain root telling AI models who you are, what you pay, what your culture is like. Think robots.txt but for your reputation.
- JSON-LD structured data — Schema.org markup on your job listings and careers page. Organization type, salary ranges, locations, benefits — in a format AI trusts.
- Open crawl access — If your robots.txt blocks GPTBot, ClaudeBot, and PerplexityBot, you're invisible to the primary discovery channel for candidates.
This isn't marketing. It's infrastructure. And 91% of UK employers don't have it.
2. Content that AI actually cites
AI models have a citation hierarchy. They prefer:
- Structured data from your domain
- Authoritative first-party content (your blog, your about page)
- Wikipedia
- Everything else
Most employers have nothing in categories 1-2, which means Reddit threads from 2022 are defining their employer brand in 2026.
The fix: publish content that directly answers the questions candidates ask AI:
- "What's it like to work at [Your Company]?" → A genuine, specific culture page
- "What does [Your Company] pay?" → Published salary ranges
- "What benefits does [Your Company] offer?" → A comprehensive, up-to-date benefits page
This content doesn't need to be beautifully designed. It needs to be factual, specific, and crawlable.
3. AI visibility monitoring
You can't manage what you don't measure. Traditional employer brand metrics — application rates, review site ratings, career page traffic — don't capture the AI layer.
What you need to track:
- What does each AI model say about you? (and is it accurate?)
- What sources does AI cite? (your data or Reddit?)
- How does your AI presence compare to competitors?
- Has anything changed since last week? (AI models retrain regularly)
This is a new discipline — call it AI employer brand management, Generative Engine Optimisation for TA, or whatever you want. The name doesn't matter. The capability does.
What the careers page becomes
The careers page still has a role. It's just a different one.
Before: The careers page was a discovery tool. Candidates found you there. Now: The careers page is a conversion tool. Candidates who've already been pre-qualified by AI arrive with specific expectations — and the careers page either confirms or contradicts what AI told them.
This changes what the page needs to contain:
- Salary transparency: If AI told them you pay £65K–£80K, your job listing better say the same. Contradictions tank trust.
- Specifics over slogans: "We're a great place to work" means nothing. "3 days remote, 27 days leave, £7K pension contribution" means everything.
- Structured data in the code: Even if humans never see the JSON-LD, AI does. Every job listing should include machine-readable salary, location, and employment type.
The uncomfortable conversation
Most TA teams are still spending their budgets in the wrong places:
| Where budget goes | Why it's fading |
|---|---|
| Careers page redesigns | 60% of candidates never click through |
| Job board premium listings | AI provides the same information for free |
| Review site management | AI doesn't cite these platforms |
| Employer brand videos | AI can't watch videos (yet) |
| Where budget should go | Why it matters |
|---|---|
| Structured data implementation | AI trusts machine-readable data most |
| AI visibility monitoring | You can't fix what you can't see |
| Content that answers candidate questions | Becomes AI citation source |
| Salary transparency | Biggest single factor in AI accuracy |
This isn't about abandoning traditional recruitment marketing. It's about recognising that the primary candidate touchpoint has shifted — and allocating accordingly.
The window
Right now, AI employer visibility is a genuine competitive advantage. Virtually no one is doing this well. The average UK employer scores 34/100 on AI visibility.
In 12-18 months, this will be table stakes. The employers who start now will have accumulated data, established authority, and optimised their AI presence while competitors are still debating whether to add an llms.txt file.
The careers page isn't dead. But the era of the careers page as your employer brand's front door? That's over.
See what AI says about you
Run a free audit. 30 seconds. See exactly what candidates hear when they ask AI about your company.
→ Run your free AI employer brand audit
Sources: SparkToro (2024), Pew Research (2025), OpenAI (2025), PerceptionX (2025), Meta Investors (Q1 2025), OpenRole audit data — 500 UK employers (2026).