That's not a search problem.
It's a process problem.

A shared inbox and good intentions.

Hiring is the only system in a business that touches revenue, brand and culture at the same time. It decides who does everything else, including the engineering, the marketing and the sales. It costs more per decision than most product launches.

It gets a shared inbox and good intentions.

No design before launch: the role goes to market on a paragraph borrowed from the last ad. No owner: hiring outcomes sit with HR, the hiring manager, finance and the founder all at once, which is to say nobody. No number: ask what a declined offer costs and the room goes quiet. No alignment: the people who will judge the candidates have never compared notes on what they are judging.

So when a hire does not work out, the inquest examines the search, because that is where the failure became visible. The first week, where most of these hires are actually lost, leaves no trace, so nobody goes back to it. That is the whole problem, and it is a fixable one.

A hire is not one job. It is three.

They are the same three the business already pays to align.

• A product. What the candidate is buying. The role, the manager, the money, the first project, the story about why this seat matters.
• Marketing. How you reach them. The ad, the network, the agency, the reputation that arrives before your first call does.
• Sales. How you qualify and close. The screen, the interviews, the offer, the negotiation, the week between the final interview and the yes.

In most hires, the founder sells one job in the first conversation. The ad describes a second. The panel interviews for a third. The offer letter prices a fourth. The candidate is the only person who sees all four, and they draw their own conclusion.

If product, marketing and sales did that to a customer it would be a board-level incident. Ask what the business spends keeping the founder, the hiring manager and the panel aligned on one role, and the honest answer is usually nothing. The good news is that alignment is a skill the business already has; it has simply never been pointed at a hire.

Each stage inherits whatever the one before it produced.

Every hire runs through the same seven stages in the same order. The businesses that hire well have designed each one; most businesses design one. Here is what good looks like at each stage, and where it usually drifts.

• The brief. Everyone who will judge the hire has agreed what the role has to change in the business and what good looks like a year in, on one page, before the ad. In practice most roles go to market on a duties list adapted from the last one, and everything downstream inherits the fog.
• The search. Where the people come from, chosen on purpose, run by someone who has read the brief. This is the stage everyone designs, because it is the one that can be handed to somebody else. A well-run search pointed at an undefined target fails with great efficiency.
• The shortlist. One person owns it and reads every application against the same standard. Usually it is whoever had time, against markers that stopped meaning anything once every application was written by the same tool.
• The interviews. Each interviewer has a lane, and answers are scored against something written down before anyone walks in. Three interviewers returning three verdicts is not healthy debate; it is three private standards, and the most senior impression wins the debrief.
• The decision. One named owner, the evidence on the table, made within a day of the last interview. Ask most panels who decides and you get four names.
• The offer. The number, the flexibility and the approver are settled before the final round, so the yes moves faster than the counter-offer. The strongest candidate decides at their pace, and their pace is faster than most approval chains.
• The first ninety days. Someone checks, at thirty days and at ninety, that the job the person is doing is the job they were hired for. Almost nobody does, so a quiet exit eighteen months later gets recorded as a bad hire rather than a bad brief.

Engineering has a name for a system where a bad input at the first stage is processed faithfully by every stage after it, and would never ship it. Hiring ships it every month. Get the front of the process right and everything behind it gets faster and easier to defend.

Five things being blamed this month, and the stage that actually owns them.

• Ask your last five hires whether the job matched the ad. Most people who take a job say it turned out to be a different one, and they decide whether to stay in the first few months. That's not a retention problem. It's a process problem, and the stage is the brief.
• The candidate accepted, the recruiter closed the file, and nobody spoke to them for five weeks. They started somewhere else on the Monday. That's not a loyalty problem. It's a process problem, and the stage is the offer.
• A good share of your new starters came in above the band because the number was decided in a hurry at the end of a long search, and the premium compounds through every pay rise after it. That's not a salary problem. It's a process problem, and the stage is the offer.
• The process got longer, the interviews got fewer, and the candidates say nobody heard them. More roles are staying open than a year ago and the explanation in the leadership meeting is the market. That's not a market problem. It's a process problem, and the stage is the decision.
• Seven candidates met, four offers made, four declined, and the call that reaches the leader says nobody has had any candidates. The fixes that follow, a second agency and a wider net, are aimed at supply. That's not a search problem. It's a process problem, and the stage is the offer.

Give hiring the full show. It costs an hour.

Write the brief before the ad. Agree the scorecard before the first interview. Name the owner before the panel meets. Decide the offer before the candidate does. Check at ninety days that the job is the job. Structured briefs, scored decisions, evidence over opinion: the discipline the business already applies to a launch, applied to a hire.

And before changing anything, count. How many met, how many offered, how many said yes, and the stage where the good ones left. The count shows the stage, and the stage is where the next hire gets easier.

The Talent Audit does that count for the whole system: it measures how you think you hire against how you actually do, stage by stage, with a cost on each gap and a first move against each one, so the next fix is aimed at the stage that is actually drifting.