Your Employees Are Shaping What AI Says About Your Brand
I have been thinking a lot about how to help organisations approach managing narratives and their corporate brand reputation in AI summaries. It is true that in the last eighteen months, companies have woken up to the realisation that when a customer, a journalist or an investor asks an AI system about their brand, the answer matters. Boards are realising that their corporate brand narrative is being rewritten by AI. AEO programmes have been commissioned. Content strategies have been rebuilt. Teams are auditing what the models say and are working on shaping the underlying inputs to shape the AI output. All sensible actions.
Brand reputation in AI answers is a real commercial question. It’s often the first impression anyone will get of your business. And a true litmus test for any brand’s communication strategy, storytelling capability and ability to control narrative.
From an employer brand perspective, this opens an interesting opportunity: When a candidate asks an AI assistant what it is like to work at a company, or when an investor asks how the workforce feels about a leadership transition or new strategic directive, the answer is not assembled from press releases or investor decks.
It is assembled, primarily, from employee voices in public. Conversations on Reddit and LinkedIn, and the wider commentary that echoes what employees post on review sites like Glassdoor and Indeed. It is the accumulated public record of what people said about their employer that sits completely outside of what may get captured in an employee survey.
That is data. It’s being built there, feeding the answers you have started investing in to get right everywhere else. And for many organisations, it is still a blind spot.
Why this matters now
Employer brand was, for a long time, a respectable line item with limited executive urgency. Three things collided in 2025 and 2026 to change that fundamentally.
The talent market is structurally short. ManpowerGroup's 2026 data puts 72% of employers reporting difficulty filling roles. The World Economic Forum expects nearly 40% of core job skills to change by 2030. In this environment, how a company is perceived as an employer is not a recruitment-marketing question. It is a growth constraint. One your investors and competitors are watching.
The workforce is quietly detached. Gallup's running measurement of manager engagement has it at 22%, down from 30%. The gap between what organisations believe about their culture and what employees feel continues to widen.
AI arrived and changed how we research. AI builds your employer brand narrative for everyone to see. The content it draws on most heavily is not managed. Research published by Ahrefs in 2025, covering 75,000 brands, found that branded web mentions correlate with AI visibility at roughly three times the strength of backlinks: a correlation of 0.664 versus 0.218. Semrush's 2026 index, drawing on over 126 million AI search prompts, has corroborated the same basic direction. Employee reviews are a dense, high-trust element of exactly those branded mentions. They carry the language of lived experience, they are specific to named employers, and the models treat them accordingly - with due importance and value.
At a roadshow we ran together with Bowen and Craggs earlier this year, we shared research that suggests that most LLM brand visibility is driven by long-term brand equity rather than recent SEO signals. The methodology warrants scrutiny, and 63% should be taken as directional rather than definitive. But the direction itself is unmistakable: the persistent, accumulated body of what people say about a brand over time is doing most of the work. For the employer brand, that body of your own content and the employee voice in social and review platforms.
The voice AI trusts most
It is worth pausing on which voice AI tends to weigh most heavily, because it is the same voice people have always trusted most. Edelman's 2026 Trust Barometer finds that “my employer” is the most trusted institution among employees, at 78%. That is higher than government, higher than media, higher than most categories of business communication. People trust what other employees say about a workplace. And so does AI.
The stakes are bigger than ‘just’ a bad review
A poor Glassdoor rating has always been uncomfortable. But the consequences of an unmanaged employee voice data now run wider than recruitment.
On talent: when 72% of employers are already struggling to hire and the skills gap is growing, your employer brand is directly connected to whether you can execute your strategy. Access to talent is a constraint. How you are perceived as an employer in the AI answers a candidate sees before they ever visit your careers page, impacts your ability to hire the right people and is part of your growth story.
On investor trust: human capital is increasingly material to how boards and institutional investors assess a company's durability. Aon, Deloitte, and J.P. Morgan have each published work on the relationship between workforce health and long-term value creation. I am not suggesting that an AI answer about your employer brand moves your share price. But the underlying workforce data it reflects, which is the same data investors are increasingly asking about, does.
On the broader cultural conversation: employer brand is part of the corporate brand. When a journalist is building a story about a company's culture, or when a public debate breaks out about pay or conditions in a sector, the public narrative feeds from the same employee-voice data that feeds AI answers. These are not separate problems. You cannot protect your corporate brand from scrutiny by ignoring the employer brand element.
The old model has run its course
Employer branding that relies primarily on purpose narratives and culture statements to substitute for material fairness has lost most of its leverage with employees. The workshop-produced EVP, the values poster, the “we're a family” messaging: these are not landing in the way they once did. Employees are more sceptical, and that scepticism is legible in the public record.
What you need now is evidence. Not a claim about culture, but a demonstration of what the data shows, where the genuine strengths are, where the gaps between promise and experience exist, and what the organisation is doing about them.
What to do about all this
The starting point is reading the part of your brand story you are not managing. That means looking at what your employees are saying publicly, not as an occasional survey exercise, but systematically: clustered by theme, scored by sentiment, segmented by function and region, and benchmarked against the specific companies you compete with for talent.
In one engagement with a global FMCG multinational, that analysis revealed a retention risk concentrated in a single discipline whose workload sentiment was critically negative and well below every other function, a data point that would have otherwise remained hidden in a company-wide average. The named competitors provided the benchmark that made the finding actionable.
In a second engagement with a major UK retailer, connecting the external public record to the company's own internal data allowed us to rebuild the employee value proposition on evidence rather than workshop consensus. The framework was signed off at executive level because its strength was rooted in lived experience translated into data.
In both cases, the external intelligence connected to what the organisation already collected internally to deliver a value that previously didn’t exist in this form: a 360 perspective of their employer brand built on employee inputs and data.
Get in touch with IDX if you'd like to hear more about how we are helping our current clients.
Let's chat
Whether you're looking for service, support or a future strategic partner - we're here to help.
LDN