The second Talking Point: the gloss, the varnish, and the rejection email nobody reads

Two episodes in, and I am already having to rethink what The Talking Point is for. I thought it was a place to explain and champion things. After Monday, I think it is a place where people admit things.

Episode two was about candidate trust, and it went somewhere I did not expect. Thank yous first, because they are the point.

The best employer brands don’t manufacture stories. They uncover what’s already true. Sell The Truth: Leadership Storytelling For Cultures That Win by Bryan Adams shows leaders how to turn truth into stories that build belief, trust and competitive advantage.

Order now

Thank you, Steven and Kellie

Steven Brand is Global Head of Employer Brand and Attraction at Rolls-Royce. Kellie McCann has spent thirteen years in employer brand across H&M, Booking.com, Amazon, Walmart and Robinhood, and is now working in Global Employer Brand at Radiology Partners. Different sides of the Atlantic, different industries, same problem.

We keep tracking to a minimum. The video is hosted on YouTube, which may place cookies that measure viewing and personalize content. We only load the player after you consent to these cookies. You can update your cookie settings any time.

Both of them paused early on to say they were speaking personally rather than for their employers, and I want to flag that rather than bury it, because it is exactly what made the hour worth having. What followed was not a panel of people defending their organisations.

What we also had to do is take stock of wider society. Whilst I opened asking whether we are responsible for the trust erosion, through collective mass of 'meh' content, oversold promises and bad company signals; we also reflected that the world teaches us to be cynical, asks us to question everything, to disbelieve leadership, and each post in our social media feeds raise the "do even believe this is real?" sentiment. What chance do positive employer brand stories have?

Steven went first, and went further than I expected. His argument is that a lot of employer brand work, his own included, has carried a layer of gloss or varnish. Not lies, but the version of the truth most likely to get signed off, because the job has been measured on applications rather than on honesty. "It's marketing," he said. "It's not necessarily just telling the truth." Anyone who has ever watched a genuinely good story get sanded down in a stakeholder review will recognise it. We have written about the sharper end of this in employer branding has a catfishing problem.

Then the turn. Steven's point is that the brief has changed under us. His organisation does not need more candidates. It needs the right people in front of hiring managers, and everyone else persuaded to look elsewhere. Which sounds obvious until you notice how few of us are measured that way. It is the same argument we made in employer branding lost control of the narrative because it stopped helping people choose.

Kellie's addition was the line I have not stopped thinking about. Trust is earned through evidence, not through pretty messaging. Candidates now triangulate their own version of your employer brand from Blind, Reddit, Glassdoor and whatever an LLM tells them in ten seconds, and that signal outweighs your owned channels because it carries more volume and more corroboration. The hiring journey does not start at your careers site any more. It starts with a question asked somewhere you do not control. If that sounds familiar, it is the flipside of what Martin Lehnert told us about GEO two weeks ago, and it pairs with our Glassdoor statistics for 2026.

Three things I am taking away

Your rejection email is a brand asset, and nobody treats it like one. Kellie called it one of the most important pieces of literature you own, and the one she sees people care least about. A template rejection twenty-four hours after applying reads as proof that nobody looked. You have not only rejected a candidate, you have rejected a customer. Steven added the long tail version: a candidate without the right skills today may have them in three, five or ten years, and a bad first experience removes that possibility for good.

Application volume serves nobody, and we built it. When the maths is twenty applications for two or three interviews, effort per application collapses, one-click apply lowers the time investment again, and the volume climbs. Recruiters cannot service it, hiring slows, and thousands of people invest hope in an outcome that was never real. Our piece on LinkedIn's job marketplace and what it tells talent leaders about trust covers the same ground from the platform side.

Employer branding works better when people share what they know. EBClub brings practitioners together to exchange ideas, compare experiences, and learn from the people doing the work.

Explore EBClub

The technology did not debias anything. We spent a while on the US collective action alleging that an automated applicant screening system disadvantaged candidates over forty, women and people of colour. Those allegations have not been ruled on, and I want to be careful about that, but the wider point stands and Steven put it well: the tech was meant to debias humans, and it is proving no better at it, because it learned from human decisions in the first place. The AI cannot discern who the best candidate really is, and there's an assertion that it could even be machine-learning repetitive rejection. The workable answer is technology for volume with a human sense-checking intent. We have covered the case in hiring algorithms are having their day in court, the wider data in AI in hiring statistics 2026, and the age dimension in too young to hire, too experienced to hire.

On solutions, Kellie made the case for employer brand as reputation analyst, risk manager and comms partner rather than the team that makes the careers site, and for generating enough honest signal, employee stories, video, earned and owned, that the machines have something true to pick up. Steven made the case for advocacy through people's own networks, and for nurturing specific talent pools eighteen to thirty-six months before the requisition exists, so candidates opt themselves in or out with their eyes open. If you want the practical version of the first, start with our guide to content candidates actually trust and employee generated content.

And both of them, separately, when I asked "do we just plough on, believing in our work in Employer Brand, given how little of societal influence we can control?" -landed on the same potential unglamorous, but absolutely honest answer. Decide exactly who you are, and repeat it. Steven: "It's who we are, no matter what you say or throw against us." Kellie's addition, which I think is the even more useful half, is that you also have to know the counter-narratives as well as you know your own story, and prepare your recruiters for them. Plenty of brands let the counter-narrative become their whole public perception by never engaging with it. Worth reading alongside are employer brands confusing visibility with value.

Thank you, everyone in the room

Chloe asked two really great questions. The first was whether an ATS could send category-based rejection reasons rather than the generic template, so candidates get something, even if it is not personal. Steven's answer was yes in principle, with a warning that five categories reads as vanilla and enough categories to be useful becomes a small cottage industry. His better idea was pulling assessment tools earlier, so an unsuccessful candidate at least walks away understanding their gaps. Her second question, on whether nurture should be broad or specific, got the most honest answer of the session: specific works far better, and Rolls-Royce will need more resource to do it at the scale they want.

EB Club is growing, fast

When I wrote this up two weeks ago there were around 230 people inside the Community Hub. There are now approaching 300, with another 130 working through activation. The hub opened at the start of August.

The in-person event programme is moving just as quickly. Bristol in October. Dublin, Paris and Rotterdam in November. London hopefully at the start of December, where I have more or less pigeonholed Steven onto a panel. Next year we are going to America, five or six cities, plus Singapore, plus more of Europe. Kellie has already lodged a bid for Seattle, which is how I have learned that running a community is a bit like being in a band, and everyone wants to know why you never play their city.

Come and join us

If you are in-house, in employer branding, TA or HR, EB Club is free and built specifically for you. It is an application, reviewed within 48 hours, and the practitioner space is kept peer-only and protected, which is the entire point. Providers and partners are welcome through their own routes, just not in the peer rooms.

Join EB Club here.

Or, if you want the longer argument, the employer brand job need never be lonely ever again.

What is coming up

In two weeks we are doing humour in employer brand with Ali Spangler from Manscaped in Seattle, who have earned the right to that subject more than most. I am adding a UK voice to that one as well, because I suspect the transatlantic gap on what counts as funny is going to be half the conversation.

Then at the start of November, Heather Walker from SAP joins me on EB leadership in an AI world. How the role changes, and what we are now expected to lead, manage and measure.

Steven signed off with "fixed trust". We did not fix trust. But an hour where two senior practitioners said out loud that the gloss has been part of the job, and that the rejection email matters more than the campaign, is a better start than most of us manage.

Thanks for being there. See you in two weeks.

Steve

Takeaways

Why don't candidates trust employer brands?

Two reasons, according to this session. The first is societal: people arrive at job seeking already trained not to trust what they read online, and that cynicism predates the hiring process. The second is self-inflicted. Steven Brand's position is that employer brand work has historically been measured on application volume, which rewards presenting the organisation at its most polished rather than at its most honest.

What rebuilds candidate trust?

Evidence rather than messaging. Kellie McCann's argument is that candidates now triangulate their own version of an employer's story from Blind, Reddit, Glassdoor and AI-generated answers, and that this signal outweighs owned channels because it carries more volume and more corroboration. The response is to close the gap between what is promised and what is experienced, then repeat it consistently. Steven's framing: decide exactly who you are, and repeat, repeat, repeat.

Why does application volume keep rising?

Because candidates hedge. When the perceived conversion rate is ten to twenty applications for two or three interviews, effort per application falls, one-click apply lowers the cost further, and volume climbs again. Both guests argued this serves nobody: recruiters cannot service it, time to hire lengthens, and candidates invest hope in outcomes that were never realistic.

Is your rejection email really a brand asset?

Kellie McCann's position is yes, and that it is the most neglected one. A rejected candidate is frequently also a customer, and a template rejection twenty-four hours after applying reads as evidence that nobody reviewed the application. Steven added the long-range version: a candidate without the right skills today may have them in three to ten years, and a poor first experience removes that option permanently.

What is the Workday case about?

A US collective action alleging that an applicant screening system disadvantaged candidates over forty, women, and people of colour. The allegations have not been ruled on. The wider point both guests made is that screening technology inherits the bias of the human decisions it learned from, so it debiases nothing on its own, and the workable answer is technology for volume with a human sense-checking intent.

Can candidates get better rejection feedback?

Steven Brand's practical answer is category-based templates rather than individual feedback, which is unrealistic at volume, plus moving assessment tools earlier so an unsuccessful candidate still leaves with something useful about their skills and gaps. The caveat is resourcing: too few categories and it reads as vanilla, enough categories and you have built a small cottage industry.

How should employer brand teams approach talent communities?

Steven Brand described forecasting the skills, locations and role types the business will need 18 to 36 months out, and nurturing those specific pools so candidates self-select in or out before a requisition opens. Steve Ward's addition, from Castor Massink's work at Specsavers, is to resist spreading nurture across eight or ten talent groups and instead go deep on the one or two that matter most.