The nine-day shortlist, step by step

Nine working days is not a sales number, it is a schedule, and the reason it works is that nothing in it waits on anything else that could have happened earlier. This is what each of those days is for, written out so you can see where your own involvement sits — because three of the nine days depend entirely on you.

For context on why speed is worth engineering: LinkedIn data reported by CIO Dive found engineering, research and project management among the slowest roles to fill, with a median around 47 to 49 days from application to start date. Very little of that is work. Most of it is queueing.

Days one and two: the brief

One intake call, ninety minutes, with someone who will do the sourcing rather than someone who will hand it over afterwards.

We are trying to leave that call able to answer five questions: what this person will ship in their first quarter, which three or four skills are genuinely non-negotiable against which are preferences, what the band is in fully loaded terms, what overlap the role actually requires, and who decides. If any of those is unresolved we say so on the call rather than discovering it on day six when a shortlist comes back wrong.

Day two is the search plan and the scorecard, both written, both sent to you. If the scorecard does not match what you will actually assess on, this is the cheapest possible moment to find out.

Days three to five: sourcing and first contact

Outbound against the plan, plus our own network, plus people we have assessed before for comparable roles and kept in touch with. The last of those is why the timeline is achievable at all — a meaningful share of any shortlist is people already known to us, already screened once, already warm.

First contact is a written screen, not a phone call. It runs in the candidate's own hours, it filters harder than a fifteen-minute call does, and it gives us a writing sample for a role where writing is most of the communication.

By the end of day five we expect to be talking seriously to a double-digit number of people, and to have told you if we are not. A sourcing problem reported on day five is recoverable. The same problem reported on day nine is a restart.

Days six to eight: assessment and references

Technical screening against the scorecard, by someone who can read the answers rather than check them off a list.

An English assessment, because for a distributed role the question is not vocabulary but whether this person will say "I don't understand what you're asking for" to a senior person in another country, in writing, on a Tuesday.

Reference checks before the shortlist, not after the offer. This is the one deviation from standard agency practice that we would defend hardest. References taken after an offer has been made are a formality everybody involved is motivated to get through; references taken before a CV reaches you are information. It also means that when you say yes, there is nothing left to discover.

Day nine: the shortlist

Three candidates. For each one: what they will be good at, what they will need support on, their band and availability, the reference findings including the awkward parts, and our recommendation with a reason.

Three is deliberate. One looks like a foregone conclusion and invites a second opinion you do not need. Six is a research project that pushes the decision out a week. Three is the number a hiring manager can hold in their head and act on.

Where it actually slows down

Honesty about the limits of this: the nine days are ours, and they are the part we control.

What we cannot control is interview slot availability, feedback latency, and sign-off. The engagements that run long almost never run long because of sourcing. They run long because a shortlist landed on day nine, first interviews were offered for the following week, feedback came back the week after that, and a competing offer closed in between.

So the practical ask at intake is small and it matters more than anything else in this post: hold the interview slots before the brief goes out, name the decision-maker, and agree a feedback turnaround in hours rather than days. Do that and nine days to shortlist becomes a few weeks to start, against a median that runs to seven.