Employer branding is the work of making the promise you give candidates match the experience you give employees. Everything else follows from that sentence: the measures split into external ones, which shape the promise, and internal ones, which shape the experience. A company that does only the first half buys itself worse reviews faster.
The cost of the gap is measurable. Harvard Business Review put it at at least 10 % more per hire for a company with a bad reputation: the same candidate simply costs more. LinkedIn's own talent research finds the opposite effect on the other side, with a strong employer brand reducing cost per hire and turnover rather than only improving perception.
What follows is the part that is usually missing: the internal and external measure split, how to write an EVP that survives an exit interview, the four metrics worth tracking, and what to do about a bad review. If you want the definition and the basics first, our colleagues covered those in was bedeutet Employer Branding and, in English, in what is employer brand.
Internal and external measures, and why the order matters
External measures shape what candidates see before they apply. Internal measures shape what employees tell people at parties. The second group produces the raw material for the first, which is why running the campaign before fixing the experience reliably backfires: more applicants arrive, meet the same reality, and leave a review about it.
The practical rule is a ratio. If your budget is going 80 % into external measures, the campaign is running ahead of the product. What the product looks like from inside is what an employee engagement survey and an eNPS survey report on, and the values underneath it are covered in the company values guide. Reverse it for the first two quarters and the external work gets much cheaper afterwards, because the stories already exist and do not have to be invented.
| Measure | Type | Effort | Shows up in |
|---|---|---|---|
| Shorten the application to under five minutes | External | Low | Application completion rate, within weeks |
| Reply to every review, good and bad | External | Low, ongoing | Rating trend and candidate trust |
| Rewrite job ads around the work, not the wish list | External | Medium | Applicant fit, first interview drop-off |
| Fix onboarding for the first 30 days | Internal | Medium | Early turnover, and the stories new hires tell |
| Make one manager behaviour a promotion criterion | Internal | High, slow | Everything, eventually |
| Ask leavers what they will say about you | Internal | Low | The gap between promise and experience |
How clearly do you read as an employer right now?
The employer branding check shows in a few minutes how clearly and humanly your company comes across as an employer, and which lever moves most. Free, with instant AI analysis, by Belistic.
The EVP: one sentence that survives an exit interview
An employer value proposition is the deal in one sentence: what someone gets from working here, and what it costs them. Most published EVPs describe only the first half, which is why they read as recruitment copy and convince nobody who has met the company.
The test is simple and uncomfortable. Read your EVP to three people who left in the last year. If they laugh, you have a marketing statement. If they say yes, that was the deal, it just was not worth it for me
, you have a real EVP with an audience problem, which is a much better position to be in.
A usable EVP names a trade-off. Autonomy and short decision paths, with the ambiguity that comes with them.
Deep specialisation in one industry, which means less variety.
Anything that sounds purely positive is describing a company nobody has ever worked at.
Collect what people already say
Six current employees, three leavers, and every review from the last year. You are looking for the words that repeat, not for the words you like.
Separate what is true from what is aspirational
Both belong in the strategy, but only the true half belongs in the EVP. Aspirational claims are what candidates discover in month two.
Name the cost
Every real deal has one. Naming it filters out the people who would have left in six months anyway, which is the whole point.
Test it on three leavers
The exit interview is the cheapest research you will ever run. The exit interview guide has the question set and the exit interview survey has it ready to send.
Put it in the job ad before the website
The job ad is where an EVP either works or does not, and it is the cheapest place to find out.
Four metrics, not fourteen
Employer branding dashboards usually collapse under their own weight because they mix reach metrics with outcome metrics. Four numbers carry the whole story, and three of them are already in your systems.
| Metric | Where it comes from | What it tells you |
|---|---|---|
| Application completion rate | Your applicant tracking system | Whether the process is the problem. Moves within weeks. |
| Turnover in the first 12 months | HR system | Whether the promise matched the reality. The honest one. |
| Share of hires from referrals | Applicant tracking system | Whether employees recommend you when it costs them credibility. |
| Rating trend, not rating level | Kununu, Glassdoor | Direction beats absolute score, which is mostly company age. |
Referral share is the metric worth arguing for. Recommending an employer costs the recommender something if it goes wrong, which makes it the only employer branding number that cannot be bought with budget.
Bad reviews: what to do and what not to
Small and mid-sized companies: what actually works
Most employer branding advice is written for companies with a brand team. Without one, three things carry almost all of the effect, and none of them needs an agency.
The first is speed. In skilled trades and manufacturing the candidate who waits four days for an answer has already signed elsewhere, and reply time is a scheduling decision rather than a budget one. The second is specificity: a small company cannot outspend a large one on reach, but it can describe the actual job with a precision that a corporate job ad never risks. The third is the people already there, which is why referral share is on the metric list above.
Industry context matters more than general advice here. We covered the shortage mechanics for manufacturing in the workforce crisis guide and for the tooling industry specifically in skilled worker shortage in tooling, and what benefits actually move the needle in Austria in the employee bonus guide.
Employer branding is working when
Referral share is rising without a bonus programme
Leavers describe the deal the way you wrote it
First-year turnover is falling
Reviews mention specifics rather than adjectives
It is a campaign when
The budget is 80 % external
The EVP contains no cost, only benefits
Success is reported in impressions and followers
The first response to a bad review is legal
Find out why the right people are not applying yet
A few minutes, an AI analysis, and the lever with the largest effect for your situation. Free, by Belistic.
The short version
• Employer branding is making the promise outside match the experience inside. Everything else is a tactic.
• A bad reputation costs at least 10 % more per hire (HBR). The gap has a price whether you measure it or not.
• If 80 % of the budget is external, the campaign is running ahead of the product.
• An EVP names a cost, not only a benefit. Test it on three people who left.
• Four metrics: application completion, first-year turnover, referral share, rating trend. Reach is not one of them.
• Do not try to delete bad reviews. Answer them for the next two hundred readers, and treat three mentioning the same thing as one finding.
• Without a brand team, speed, specificity and your existing people carry almost the whole effect.






![Knowledge Silos: Stop Losing What Your Teams Know [2026 Guide]](https://www.teamazing.com/wp-content/uploads/2026/04/employee-knowledge-silos.jpg)
![eNPS: How to Calculate the Employee Net Promoter Score (Formula + Benchmarks) [2026]](https://www.teamazing.com/wp-content/uploads/2026/04/enps-employee-loyalty.jpg)
![360-Degree Feedback Questionnaire: Examples, Template and Process [2026]](https://www.teamazing.com/wp-content/uploads/2026/04/360-feedback.jpg)