Summary: Employer brand has no instrument for judging its own craft. Awards are paid, self-nominated, and assess a written description rather than the work, and the only people who ever encounter this work in context, while actually deciding whether to apply or whether to believe any of it, are candidates and employees. They're the one group nobody asks. If we can't tell good work from well-described work, we can't tell what helps, which means we can't measure it, staff it, or build an investment case for it. The second half of this piece sketches what we could build instead: peer review, no entry fee, and a set of compulsories that put the work side by side.

How would we even know if employer brand work is worth its weight?

Ask anyone in employer brand to name the strongest work of the last five years and you'll get answers quickly. Ask how they know, and it turns out they read about it.

Before I take that apart, let me say what I mean by craft, because the word does a lot of hard labor in this field. Craft here means work that is smart, distinctive, and aimed at a real problem. Competent execution is the floor. A clean layout built on a brief nobody really questioned, no what-problem-are-we-solving protocol anywhere in sight, is a well-made nothing, and our field produces a great deal of it.

Now, how do we decide which work is good?

We don't. That's why we have the awards shows. RAD, Rally, the Employer Brand Management Awards. They are the only organizations documenting this field at all, which is worth saying up front and worth saying again later. Not a dis. It's the field and how it formed.

Try to audit one of their verdicts and you run into a wall almost immediately. What's available to you is a description of the work, written by the people who made the work, submitted as a paid entry, then rewritten by an awards copywriter who hadn't seen the work in context either. Nobody in that chain looked at the thing while it was doing its job. Other than a careers site, there aren't many ways for someone who isn't a job seeker or an employee to see this work at all.

I juried one of these. The entries in my category were well below my version of a craft standard, across every dimension I'd measure. That presented a problem. By my scoring, none of them should have won anything. I'd give them all an A+ for effort and not for output. But the process doesn't work that way. Even if I'd scored every entry at the bottom of the scale, one of them still comes out highest, and highest gets the award. There is no floor. A category can be won by work that cleared nothing at all, and nobody reading the winners list afterward can tell the difference. How is that good for the practice?

There is exactly one group of people who encounter employer brand work in context, which is to say while actually deciding whether to apply, whether to stay, whether to believe any of it. Candidates. Job seekers. Employees. They are the only researchers with real access, and they are the one group nobody asks. Everyone doing the evaluating in this field encounters the work as a case study, stripped of the moment it was built for.

Hold onto that, because it comes back at the end and it's the thing we have to solve.

What exactly is missing?

No adjacent benchmark. Nothing that lets you say this careers site is stronger than that one on a dimension someone outside the industry would recognize.

No test. No standardized brief anyone runs the same way twice.

No comparison, because the work is never adjacent to other work. A shopper sees ten cereal boxes at once, and that involuntary adjacency is why consumer standards climb. Nobody has ever seen ten careers sites at once. Not a candidate, not a recruiter, not you. Each one is visited alone by someone who arrived with intent and leaves inside ninety seconds.

No critical discourse. Nobody reviews employer brand work. There's no equivalent of a film critic, a design critic, an architecture critic, or even the comment section under a rebrand on Brand New, which is savage and does more for standards in a week than most of what we've built. Communication Arts publishes criticism alongside its awards. Design has had that for decades. It's missing entirely from EB.

No professional organization running anonymous, unpaid, open-panel competition on the work itself. I know of none. Every route to visibility in this field runs through an entry fee first.

Don't companies already test this?

Serious employers do test this, and the objection deserves a straight answer rather than a dismissal. They run usability testing on careers sites, A/B test job posts, and track where applications fall out of the funnel. There's also CandE, the candidate experience benchmark research that's been collecting job seeker feedback across employers since 2011, at a scale most of us underuse. It's the closest thing this field has to an instrument, and it proves the hard part is solvable. You can ask candidates, at volume, and publish what they say.

What it measures is process. How long you took to respond, how you handled rejection, whether the interview felt fair. Real, and worth fixing. But nobody in that data set is being asked whether the rejection email was any good. The infrastructure exists for everything except the craft.

Which leaves internal testing carrying the rest, and internal testing has a ceiling problem. The comparison set is always the same brand's previous version of itself. You learn that variant B beat variant A. You never learn whether A and B were both far below what the category was capable of. Optimization inside a low ceiling reads as progress right up until someone shows you the ceiling.

Isn't that what the awards are for?

Industry recognition is not a measure of craft, and it can't be, for reasons built into how it operates. This is the version I'd write if I weren't worried about upsetting people. Worth reading at full strength before deciding how much of it is fair.

Entry is paid, so what's visible is filtered by budget before quality. Entry is self-nominated, so the sample is people who believed their work would win. The submission is written by the entrant, so judges assess a narrative about the work rather than the work. Categories bundle craft with results, so volume can carry a mediocre execution. Judges are drawn from the same small population being judged, which is unavoidable at this size and still a problem. And almost nothing gets published in enough detail for anyone outside the process to check the verdict.

Every one of those is defensible on its own. Together they produce a system that reliably identifies work that was entered, and nothing beyond that.

Is Cannes fairer? Is AIGA, or Graphis, or Awwwards? Mostly yes, and the reason isn't better judges or cleaner processes. Cannes has enormous entry fees too. The difference is that in those fields the work is public, so anyone can check the verdict against the thing itself. A bad Cannes call gets argued about for a year because everyone has seen the ad, the spot, the campaign. Nobody argues about a careers site verdict, because nobody else has seen the site.

The award programmes are doing what awards do everywhere. Awards were never designed to be an evaluative instrument. Asking them to be one is asking the wrong institution, and the failure I'm describing belongs to the rest of us for never building the right one.

Why does this matter?

If there's no reliable way to tell what high craft looks like, there's no reliable way to tell what helps. Then there's no reliable way to measure. No reliable way to staff, invest, prioritize, and all down the line. It's guesswork. I can't accept that.

We can't isolate which decisions produced which effects, because we can't grade the decisions. Every claim this field makes about its own value is a claim about correlation between things not sufficiently measured. A campaign ran, applications went up, and somewhere inside that gap sits the actual causal work: an insight, a piece of writing, a structural choice about who the message was for. We report the applications. We can't report the rest, so we never learn which part was carrying the heavy weight.

Which means the discipline can't accumulate. Each program starts from roughly zero, borrows whatever the last vendor pitch deck showed, and produces something that will itself be described rather than examined.

A field that can't demonstrate what its craft contributes is the easiest field in the economy to automate, because nobody can say what would be lost. If the difference between excellent and adequate has never been established, adequate at a fraction of the cost wins every budget conversation it enters.

I wrote recently about employer brand maturity as a ladder, and about how programs get funded as a cost center because craft never compounds into an asset. This is the mechanism underneath that. You can't build an investment case for something when you can't establish that X produced Y. The systems argument and the measurement argument, the two serious cases anyone makes for this work, both fall into the same hole. Not for lack of rigor. For lack of an instrument.

Am I complaining or helping?

I have a sketch, and I offer it as an attempt to reconcile, honor, and unite craft and its practitioners, rather than as an answer to what is good.

Peer-reviewed, not judged. No entry fee. The model Awwwards and CSS Design Awards use, where the community scores against public criteria and the criteria themselves are the durable output.

Start with the compulsories, because they're the part that could run next month for almost nothing.

Competitive diving has a required list. Every diver performs the same set, and you watch them one after another. Follow that model. Every employer brand program on earth produces the same small set of components: a job posting, a careers landing page, a display ad, a LinkedIn About Us block, a rejection email. Set the brief, collect every version, show them side by side. Free form categories can exist alongside it if we want them. The compulsories rebuild the shopping aisle artificially and require no new work from anyone, because the work already exists.

Line up fifty rejection emails, and the memorable one won't be the one with better typography. It'll be the one somebody actually thought about. It'll be instantly obvious.

The candidate perspective problem gets solvable here, and that's the reason compulsories matter more than the rest. A rejection email takes about ten seconds to read, a LinkedIn About Us block maybe twice that. These components are small enough that you can put them in front of people who are actually job hunting and get a real reaction, at scale, cheaply. Not practitioners assessing a case study. People in the middle of the thing the work was built for. That's the only version of this I've sketched where the group with real access is in the judging process.

Then the scored events, which are harder and worth arguing about. Weight them the way gymnastics is weighted: difficulty multiplied by execution. Difficulty is the whole point. Distinctive work coming from a gaming studio is not the same accomplishment as distinctive work from a regional utility, a defense contractor, or a nursing home operator. Degree of difficulty would encode category convention, audience hostility, regulatory drag, budget, and how thoroughly a sector has trained people to expect nothing from it. Nobody medals by clearing a low bar cleanly.

I'm not going to pretend I know how to calibrate that. The calibration is the argument, and it should happen in public.

I've cut the third event I was going to propose. I'd sketched a measured category won purely on the numbers: conversion, offer acceptance, twelve-month retention. It doesn't survive my own argument. Those are the same correlations I spent this piece saying can't isolate craft, and a competition that rewards them would reward the employer with the best labor market, not the best work. Pay the best and you get speed to hire.

And the panel opens. A closed judge pool drawn from the same small population is one of the structural failures I listed above, so stop drawing from it. Let designers, copywriters, UX people, anyone who wants a say, have one. They all have careers, which means in a manner of speaking the whole world has a stake in this.

The criteria are the proof of craft standard. Publish them, argue about them publicly, before anyone competes. Peer review keeps the argument permanent and free, which is precisely what entry fees have always quietly prevented.

A field that can't tell good work from well-described work will keep producing the second kind and keep wondering why nobody outside it cares, and why leaders cringe having to fund it. We don't currently know whether we're high craft or low, whether we're solving real problems or the ones easiest to get approved, and we have built nothing capable of telling us. Every argument we make for our own value rests on that gap.