Your EVP & Employer Brand Aren't Being Read. They're Being Scraped.

Go and ask your LLM of choice what it's like to work at your company. Right now. I'll wait... ⏱

Done? Good. Because whatever it just told you - that's your employer brand for a growing number of candidates. Not the EVP document you spent six months crafting. Not the careers page your agency designed. Not the brand film that won an award at some conference. The three sentences an AI just generated from whatever signals it could find about you.

That should worry you. And if it doesn't, you're not paying enough attention to where this is heading.

I'm watching the future take shape

I called my business 'Reveal The Real' and that's not just a name it's part of my moral compass. So, I'd feel very uncomfortable if I wasn't completely transparent here - I'm not the expert on AI visibility auditing. Far from it in fact! But I'm close enough to the people who are that I'm getting a front-row seat to how this is going to reshape culture communications, EVPs and employer branding - and honestly, it's fascinating. 🤓

I'm currently working alongside a specialist vendor and an enterprise client on what I can only describe as an AI employer brand audit. The client is a major global organisation - one you'd recognise - and they're taking this seriously enough to have run over 1,600 structured prompts across six different AI tools, three candidate personas, and six regions. This isn't someone having a casual play with ChatGPT. This is a controlled experiment designed to answer a specific question, "how does AI understand us as an employer, and how does that compare to our competitors?"

Sitting in the middle of the email threads between the parties I've connected, watching the methodology take shape, I'm learning terminology and concepts that I think every culture, EVP and employer brand professional needs to understand. So let me translate what I'm seeing in a way that's helped my simple mind get to grips with it.

Three concepts you need to know

Visibility frequency is the simplest one. When a candidate asks an AI tool a non-branded question - something like "which companies should I consider for a career in advertising?" or "who are the best employers in tech?" - does your company even get mentioned? How often? And where do you rank compared to your competitors? This is measurable. You can run the same question across multiple AI tools, repeat it dozens of times, and start to see patterns. Some companies show up consistently. Others are invisible.

Ranking dynamics follows naturally. When AI does mention your company alongside competitors, where do you sit in the pecking order? And is that ranking stable, or does it shift depending on which AI tool you ask, which region the question implies, or which candidate persona is asking? If ChatGPT ranks you first but Gemini puts you fourth whilst Claude has you in eighth, that tells you something about how consistent your signals are across the sources those models are drawing from.

Narrative compression is the one that really made go, "Wait. What?". Think about it this way. Your EVP is probably a detailed document - themes, messaging, proof points, maybe 20 pages of carefully chosen language. When AI answers a question about your company, it takes everything it has ever absorbed about you - job descriptions, Glassdoor reviews, news articles, career site copy, LinkedIn posts, Reddit threads, employee testimonials - and squashes it all down into a handful of sentences.

That squashing is narrative compression. Your entire employer brand, reduced to a few words.

The question then must be, "what survived the squeeze?" Did AI compress you into something distinctive and accurate? Or did it compress you into the same bland mush it says about everyone else - "collaborative culture, opportunities for growth, commitment to innovation"? 🥱

And, for me, this is where it gets properly interesting. When you test this at scale across multiple AI tools, you start to see whether your brand compresses consistently or whether it fragments. Maybe ChatGPT says one thing about you, Gemini says something different, and Perplexity says something else entirely. That fragmentation - the coherence gap - means your signals are contradictory enough that different AI models are reaching different conclusions about who you are as an employer.

Why this matters more than your next campaign

Charlotte Marshall wrote recently about employer brand teams over investing in low-frequency content - the brand films, the launch campaigns, the big moments - and underinvesting in the high-volume signals that AI actually learns from. Job descriptions. Recruiter scripts. Career site copy. Review site themes. She's right. AI doesn't build its understanding of your company from the one campaign you loved. It builds it from the signals that appear over and over again.

Will Innes is talking about "agile EVP" and building specialised AI agents trained on decades of employer brand IP. He's pointing at something important too - the gap between prompting a chatbot to rewrite a job ad and actually building intelligent systems that connect research, strategy, and activation at pace.

What I'm seeing from the inside of a live engagement is what happens when you take these ideas seriously and apply proper rigour. When you treat AI-mediated employer brand discovery as a measurable channel - not a novelty, not a future problem, but something you can audit, benchmark, and act on right now.

What you can do about it

Start with the simple exercise I opened with. Ask multiple AI tools what it's like to work at your company. Ask them who they'd recommend in your sector. Note what comes back. Is it distinctive? Is it consistent? Is it even accurate? And if you're a large, international company where different countries are doing their own thing because they 'know what's best locally and to hell with the centre of excellence' - I can probably save you the exercise. The answer is a big, fat no. 😬

That will give you a rough sense of the picture. But if you want proper diagnostic insight - the kind that can genuinely inform your strategy - you need a structured approach. Multiple AI tools, controlled prompts, repeated runs, competitor comparison, regional and persona variation. 🔬

That's the work I'm involved in right now. And if this is something you want to explore for your own organisation, get in touch. I can connect you with the specialist expertise to run this properly and deliver insights that will actually shape what you do next.

Because your culture, EVP and employer brand aren't just documents for your leaders and teams anymore. They're training data. And right now, I'm willing to bet most people reading this don't have a Scooby Doo what the hell the AI is learning about your organisation.