Why every EVP you write ends up saying the same thing

The generic EVP is a symptom, not a writing problem

You can tell when an EVP was written from a template, because it would be equally true of any competitor. Swap the logo and nothing breaks.

That happens when the drafting starts from a framework rather than from evidence. A framework asks what your pillars are, and in the absence of data you answer with what any decent employer would claim. The output is fluent, defensible and useless.

The fix is not a better framework. It is having something specific to say before you start writing.

Where the specifics actually live

All of this already exists inside most clients. It is just not gathered, and it is never gathered in time for the EVP.

• Why good people joined. Not what the offer letter said. The actual reason they picked this over the other one. Usually a person, a specific problem, or flexibility nobody advertised.
• Why good people left. The real reason, not the exit interview answer. This is the most useful and least collected input, and it tells you what the EVP must not overclaim.
• What the founder says off-script. How they describe the company to a friend. That version is specific and slightly unflattering, which is exactly what makes it credible.
• What the work is actually like. The Tuesday, not the mission statement. Small team, no process, everyone does everything, is a real proposition for some people and a warning for others. Both are useful.
• Who has thrived and who has not. Patterns across a handful of people say more than any survey, and any manager can list them from memory in ten minutes.

A credible EVP says who it is not for

The single strongest test of an EVP is whether it would put the wrong candidate off. If it would not, it is not a proposition, it is an advertisement.

Saying the work is unstructured, or the team is small enough that everyone does their own admin, or that progression is real but slow because nobody leaves, filters better than any claim about culture. It also survives contact with the first week, which is when a generic EVP falls apart and becomes a retention problem.

This is usually the hardest part to get a client to agree to, and it is the part that makes the rest work.

The workflow that produces a specific one

The intake is the work. The drafting is fast once you have material, which is the opposite of how most EVP projects are scheduled.

• 1. Gather the evidence first. Joiner reasons, leaver reasons, the founder's unguarded description, and the thrive-and-struggle patterns. Half a dozen short conversations, not a survey.
• 2. Draft from your own previous EVPs as reference. Not as templates. What you are reusing is structure and register, never the claims, because the claims are what must be specific to this client.
• 3. Write the disqualifier before the pitch. Who this is genuinely not for. Everything else gets sharper once that sentence exists.
• 4. Check it against the handbook. The EVP promises and the policies have to agree. An EVP claiming flexibility next to a policy requiring five days in the office is worse than no EVP.
• 5. Keep the evidence with the client's standing record. Next year's refresh starts from what you gathered rather than from another blank page. This is what stops it being a rebuild every time.

Why it belongs next to the handbook

The EVP and the employee handbook are drafted from the same intake and then usually maintained as if they are unrelated documents.

They are the same promise in two registers. The EVP says what it is like to work here. The handbook says what the rules are. When they drift apart, the handbook wins in practice and the EVP becomes something the recruiting page says.

Building them from one source of client facts, in the same folder, is most of what keeps them consistent as both change.

What this does not do

It does not work without access to people. The evidence comes from conversations, and a client who will not make anyone available will get a generic EVP no matter what process you run.

It does not make a bad employer attractive. It makes an employer legible, which helps the right people self-select and helps the wrong ones opt out. If the underlying reality is bad, an honest EVP will say so and that is the correct outcome.

It is not faster on the first client. The gathering takes real time. What compounds is the structure and the register, so the second and third go quicker.

Questions