AI Has Industrialised Word of Mouth. Is Your Employer Brand Ready?

For years, employer brand teams have worried about word of mouth. What employees say to friends. What former employees tell colleagues. What candidates share after an interview. What appears on Glassdoor, Reddit, LinkedIn and other forums.

The difference now? AI can take all of that, combine it with thousands of other signals, and turn it into an answer for a candidate within seconds.

Word of mouth has been industrialised.

Consider how someone traditionally researched a potential employer. They might visit the careers site. Search Google. Read employee reviews. Look through LinkedIn. Ask someone they know who works there. They had to gather the information and form their own view.

AI changes that behaviour. A candidate can now ask a single question and receive a synthesised answer.

• Which employers offer the best career progression for someone working in finance in London?
• I'm considering jobs at Company A and Company B. Which has the better culture, development opportunities and work-life balance?
• Based on what employees say online, what are the downsides of working at Company X?

It finds signals from multiple sources, weighs them, synthesises them and presents a recommendation. Sometimes it will even tell the candidate which employer it thinks they should choose.

That is a very different candidate experience.

This creates a problem for the traditional model of employer branding. You can control your careers site. You can control your EVP. You can control your recruitment campaigns. You cannot control everything AI learns about you.

AI can draw on careers content, employee reviews, news coverage, social content, forums, awards, corporate pages, job advertisements and other third-party sources. It can also compare those signals with the same information about your competitors.

That last point matters. Candidates aren't always asking what it is like to work at Company X. Increasingly, the more important question is which companies should I consider. You may not even appear in the answer.

Traditional word of mouth had natural limits. One employee might tell five friends about their experience. A candidate might tell their network about a poor recruitment process. Someone might write a review that gets seen by hundreds or thousands of people.

AI removes much of that friction.

• One employee review can become part of an answer.
• Ten similar reviews can become a pattern.
• A Reddit discussion can become evidence.
• An old news story can influence a current recommendation.
• A competitor's clearer evidence around career development can help them get recommended ahead of you.

AI doesn't need the candidate to find each of those sources. It does that work for them. That is why I think 'industrialised word of mouth' is a useful way to understand what is happening.

The individual voices haven't disappeared. AI has become the layer that collects, interprets and redistributes them.

This may be one of the biggest changes for employer brand teams.

Research from Pew found that when Google displayed an AI summary, users clicked a traditional search result on just 8% of visits, compared with 15% when no AI summary appeared. Only 1% clicked a source contained within the AI summary itself. More recent Pew research found that 60% of US adults say they read AI summaries at the top of search results.

The implication for employer brand is significant. A candidate can learn about you without reaching your careers site. They can form an impression without reading your EVP. They can compare you with competitors without opening either company's website. They can even decide whether you belong on their shortlist before you know they exist.

For years, employer brand teams have asked what do we want candidates to know about us. That still matters. But there is now another question: what does AI actually tell candidates about us?

And then several more.

• Does AI recommend you?
• Which competitors does it recommend instead?
• What does it say you are good at?
• What does it say you are poor at?
• Which sources shape those answers?
• Does the answer match the employer brand you have invested in?
• Does that answer change depending on the AI engine, candidate priority or talent market?

This doesn't mean changing your EVP every time ChatGPT produces an answer you don't like. AI results move. Different engines use different sources. Recommendations can vary by question, geography and candidate priority. One scan is a snapshot, not a strategy.

The first step is measurement. Track what AI says over time. Look for recurring patterns. Understand where competitors consistently outperform you. Identify where AI has strong evidence for your claims and where the evidence is weak, contradictory or missing. Then decide what deserves action.

Because employer brand teams now have two audiences to understand. The humans making career decisions. And the AI systems increasingly helping them make those decisions.

Word of mouth used to travel from person to person. Now AI can collect it, synthesise it, compare it and deliver it on demand. The question for employers is no longer whether AI has an opinion about you. It's whether you know what that opinion is.