The recruitment process is the sequence that takes you from "we need someone" to "they started, and they are still here in a year". Written down it looks obvious. In practice most of it runs on habit, and the parts that fail are consistent across almost every employer: the requirement was never agreed, the panel could not find a time, and nobody would make the decision.
This is the end-to-end view. Each stage links to a deeper piece.
Stage 1. Define the role
Confirm the role is the right answer to the problem, agree what good looks like at twelve months, write the must-haves honestly, and settle the budget before anything goes public.
Where time is lost: the requirement is assumed rather than agreed. Six weeks later the hiring manager rejects a shortlist that matched the brief, because the brief was never really the brief.
Covered in planning the hire and writing the job description.
Stage 2. Attract
Publish to your careers page and out to the boards, work your existing talent pool, ask your team, and approach directly for anything specialised.
Where time is lost: posting and waiting on roles where supply is thin. Two weeks of silence is data, and the response is to change the approach rather than to wait another two weeks.
Your talent pool is the cheapest source you have and the most neglected. People who applied for an adjacent role six months ago already know who you are.
Stage 3. Screen
Fixed criteria written before you open the pile, a consistent first pass, and a recorded reason for every rejection.
Where time is lost: less in the reading than in the waiting. Applications sitting unopened for a week is the single most common way to lose a strong candidate, and you never find out it happened.
Covered in screening.
Stage 4. Assess
Whatever establishes the things a conversation cannot: a work sample, a skills test, a structured interview, a psychometric where it is genuinely relevant.
Where time is lost: assessment bolted on without a decision rule. If you have not agreed in advance what a given result means for progression, the result becomes another opinion in the room rather than a decision input.
Match the assessment to the question. Use a work sample for can-they-do-it, a structured interview for how-they-think, and a psychometric only where the trait genuinely predicts performance in that role. Covered in pre-employment assessment.
Stage 5. Interview
Two or three stages, structured, same core questions, independent scoring before discussion.
Where time is lost: panel availability, almost always. Not the interviews themselves. Put slots in diaries before the job is posted and this stage compresses dramatically.
Covered in interviewing.
Stage 6. Decide and offer
Compare against the scorecard, decide, call the same day where you can, confirm in writing immediately.
Where time is lost: the gap between the final interview and the offer. A candidate who hears nothing for ten days assumes they came second and re-engages with other processes. Even a "we are deciding this week" message holds the position.
Covered in making the offer.
Stage 7. Close and onboard
Keep contact through the notice period, get the admin done early, plan the first week before they arrive.
Where time is lost: this stage is treated as finished when the acceptance lands. It is not. The notice period is when counter-offers arrive and when enthusiasm cools.
Covered in onboarding.
The four numbers worth tracking
Most hiring metrics are vanity. These four change decisions.
Time in each stage, not total time to hire. The total tells you there is a problem. The stage breakdown tells you where it is, and it is nearly always one specific stage, usually scheduling.
Source of hire, for hires rather than applicants. A channel producing three hundred applications and no hires is a cost, not a success.
Offer acceptance rate. Below about eighty-five percent and something is wrong upstream: the pay band, the sell, or the speed. Declines are the most useful feedback you get, so ask why and record the answer.
Retention of new hires at twelve months. The only measure of whether the process actually worked. Everything else is a proxy for this. A fast, cheap process producing people who leave inside a year is not efficient.
Deliberately not on this list: cost per hire, which is easy to improve by doing worse hiring, and applications per role, which rewards volume over relevance. Reducing time-to-hire goes into this further.
Where the process usually actually breaks
Three failures account for most of it, and none of them are recruitment problems exactly.
Nobody owns the decision. The process stalls at the end because consensus was never defined. Name a decision maker before you start.
Scheduling. More elapsed time is lost to finding a mutually free hour than to any other single cause. Fix it by booking slots in advance.
The requirement was never agreed. Everything downstream inherits this, and it only surfaces at shortlist, six weeks in.
Software will not solve any of the three. An applicant tracking system makes them visible, which is a genuine prerequisite for fixing them, but the fixes are decisions rather than features.
Common questions
How long should the recruitment process take?
It depends entirely on the role, and the useful question is not the total but where the time goes. Measure per stage. If any single stage is consuming more than a third of the elapsed time, that is where to look, and it is usually scheduling.
How many interview stages are right?
Two or three for most roles. Every stage loses candidates. A fourth that exists because nobody will decide is a decision-rule problem, not an information problem.
What is the most common reason hiring fails?
The requirement was never genuinely agreed between the hiring manager and whoever ran the process. It surfaces at shortlist and costs the whole cycle.
Should you always advertise externally?
Not always. Check internally and check your talent pool first, since both are faster and cheaper. Where an internal path exists, be transparent that an internal candidate is in the running rather than running a process for appearances.
Agree the requirement properly, book interview slots before you post, keep candidates warm between stages, name who decides, and track time per stage, source of hire, offer acceptance and twelve-month retention rather than cost per hire.