The PerceptionX Visibility Index

Talent acquisition and employer brand leaders at larger organisations — our approach is built for companies of around 1,000+ employees, where reputation spans multiple markets and functions and the stakes on quality-of-hire are high.

The Employer Perception Score (EPS), built on three pillars: Sentiment (how positively or negatively you're described), Visibility (how often and where you appear when candidates research), and Relevance (whether you show up on the themes that matter, with recent, credible proof). You get a score per pillar so you can tell a visibility problem from a sentiment problem — they need very different fixes.

The Employer Perception Score is PerceptionX’s core metric that measures how positively AI describes your employer brand across themes like culture, growth, leadership, and flexibility. It blends not just visibility but aspects like sentiment, relevance and proof points into a single score you can compare against competitors.

We track the major models candidates actually use to research employers — ChatGPT (OpenAI), Claude (Anthropic), Gemini and Google AI Overviews (Google), Perplexity, and DeepSeek — rather than optimising for a single platform. Every model tells a slightly different story about your company, so we blend answers across them into one view. That's deliberate: it cancels out the quirks of any one system and gives you a more reliable read than checking ChatGPT alone. As new models reach meaningful candidate adoption, we add them.

GEO tools (Profound, Goodie, and similar) were built for marketing teams to track how a product or brand shows up in AI answers. PerceptionX is built for talent. Three differences matter:

First, scope — we go beyond "are you mentioned" (visibility) to how you're described (sentiment) and whether it counts for the themes candidates weigh (relevance). That's the Employer Perception Score, not a single visibility number.

Second, the question set — we don't track generic brand prompts. We use the real questions job seekers ask ("what's it like to work at X," "X vs Y culture"), mapped to employer-brand themes like leadership, career growth, flexibility, and wellbeing.

Third, the recommendations — GEO tools tend to push content and links. We trace AI answers back to the sources shaping them (review sites, forums, news, social) and only recommend actions that actually move the needle for talent.

No. The same employer is described very differently in London, New York, Bangalore, or Dubai, and a single global score hides that. We measure perception market by market, so you can see where your story is strong and where it's thin.

Yes — and this is where most tools stop short. How AI talks about you to a software engineer is not how it talks to you to a finance or warehouse candidate. We segment perception by function so you're not acting on an averaged-out picture. If engineering reads strongly but early-careers or commercial roles don't, you'll see exactly that, and where to fix it.

Both. AI answers are the headline, but they're assembled from somewhere. We analyse the wider landscape candidates trust — review sites, forums, social, and news — and surface which sources are doing the most to shape what AI says about you. That's what makes the recommendations actionable: you can go fix the input, not just the output.

Both, but the difference matters. You get the data and dashboards, but we partner with you on what to do with them — strategic recommendations tailored to your markets, functions, and competitors, not a self-serve report you're left to interpret. The output is a plan, not just a number.