Why your req has been open for four months

When a technical requisition sits open for four months, the explanation offered is almost always the market. Occasionally that is true. Far more often the req is open because of something the hiring team is doing, and it is one of five things.

The five bottlenecks

They are: a requisition that describes a person rather than a job, an interview loop longer than the role justifies, feedback that arrives after the candidate has accepted elsewhere, a salary band set before the market moved, and — the one that quietly causes the other four — nobody actually owning the req.

1. The requisition describes a person, not a job

Ten years of Kubernetes, deep Terraform, strong Go, ex-FAANG, and comfortable owning the on-call rotation for a platform they have not seen yet. Each line was added by a different person in the review chain, and nobody was asked to remove one.

The test is simple. Go through the requirements and mark each one as either needed in the first ninety days or learnable on the job by someone who has the first list. Most requisitions lose between a third and a half of their requirements to the second column. What is left is a job description, and it is the one that will actually get filled.

2. The interview loop has more stages than the role has responsibilities

Five rounds for a mid-level contract engineer is not rigour, it is indecision distributed across a calendar. Every additional stage adds scheduling latency, and scheduling latency is where good candidates are lost — not at the technical screen, but in the eleven days between round three and round four.

Contract roles should be decidable in two conversations. Permanent engineering roles rarely need more than four. If a stage exists because it has always existed, and nobody can name a candidate it has ever filtered out, it is not a stage, it is a habit.

3. Feedback takes longer than the candidate’s other offer

A strong senior engineer who is actively looking is typically in two or three processes at once. Exploding offers are rare; simple timing is not. If your feedback loop is five business days and your competitor’s is two, you will lose candidates you never realised you were competing for, and you will attribute it to compensation.

Agree at intake what the maximum turnaround is between a candidate finishing a stage and hearing the outcome. Then treat a breach of it as a process failure rather than a busy week.

4. The band was set before the market moved

Compensation bands are usually reviewed annually. Contract rates in scarce disciplines move quarterly. When a band is eight months old, it is not wrong by a little — it is wrong by roughly a market cycle, and the candidates who would have accepted it stopped applying some time ago.

The signal to watch is not rejection at offer. It is a screening funnel that looks healthy at the top and produces nobody who will take the call. That is a pricing problem presented as a sourcing problem.

5. Nobody owns the req

The hiring manager owns the outcome. Recruiting owns the pipeline. Neither owns the req, so when it stalls between them there is no single person whose week gets worse. Four months later, everyone involved can accurately say they did their part.

A fix that takes one meeting

Put the hiring manager, whoever runs the loop, and your recruiting partner in a room for forty-five minutes before sourcing begins. Leave with four things written down: the requirements that survived the ninety-day test, the number of interview stages and what each one decides, the maximum feedback turnaround, and the name of the person who owns the req.

That meeting is the highest-leverage forty-five minutes in the whole process, and it is the one most consistently skipped. It is also, not coincidentally, exactly what our intake call is for.