Sample frameDay 357·Hiring · Synthesis

Good morning. You've run a hiring process all year. Nobody has ever checked whether it works.

Hiring.

The Hiring Operating Review

5 min read·Apply by lunch

The question

The first morning of this elective asked how many of your last ten hires you'd make again, knowing what you know now. Here's the same question with the evidence attached: for each of them, can you point to the one thing in the process that would have told you the answer in advance? If what comes back is mostly a conversation that went well, you don't have a hiring process. You have a run of luck you've been grading as judgement.

The idea

Robert Guion spent his career at Bowling Green State University and wrote Personnel Testing in 1965, the book that put a plain demand to anyone selecting people: show that your method works. Criterion-related validity is that demand in its simplest form. A selection method earns the name only when the scores it produced actually line up with how those people later performed — which means keeping the scores, waiting a year, and going back to look. Almost nobody goes back. The scorecards go in a drawer, the interview notes are never reopened, and the process gets rated on how it felt at the time rather than on what it predicted. A second problem sits underneath the first: the only performance you ever get to observe belongs to the people you said yes to, so everyone you turned down stays permanently outside your evidence. Meanwhile hiring is run as a series of separate emergencies and judged on whether the seat got filled. Day 50 already gave you the instrument that fixes most of this — a fixed cadence at which a system gets read rather than felt, with the indicators chosen in advance so nobody can pick a flattering one afterwards; Andy Grove built Intel's management practice on the same idea in High Output Management in 1983. Point it at hiring, once a quarter, on one page.

The framework

01
Column one: every open role.
For each role still unfilled the page carries four things — the scorecard, the honest preview, how long it's been open, and where the candidates are actually coming from. Read those together and most stalled searches explain themselves. A role with no ranked outcomes is a wish list, and it attracts people who match a description rather than a result. A role sourced from your own inbox and the job boards is a role working the two thinnest seams available. A role open nine months with nothing honest published about its worst parts is a role you're still describing to yourself.
02
Which evidence actually decided it?
For every hire of the last year, write down the one thing that decided it — not the process on paper, but the moment the answer stopped being in doubt. This is Day 94's decision review run on hiring, and it returns the same uncomfortable result: the work sample and the structured scores sit neatly in the file while the deciding factor was somewhere else entirely — a conversation that went well, a referee who sounded enthusiastic, the fact that the seat had been empty four months. Record both, what the file said and what actually happened. A process you never check afterwards is a process you're merely performing.
03
Grade last year's hires against the scorecard you wrote.
Pull the scorecard for each hire whose first year is up and mark the ranked outcomes: hit, partly, missed. Call that outcome delivery and keep it as the diagnostic. The headline figure is the blunter one — how many of them you'd hire again — the number the counter-check pairs with. Expect the first pass to be uncomfortable. Most founders discover they're excellent at one class of role and poor at another, which is far more useful than a single average. Then the honest line: the only people you can grade are the ones you said yes to, so count the finalists you rejected who are visibly thriving elsewhere.
04
Column three: whose leaving would genuinely hurt?
List the names, and against each one put four things — what actually holds them here, which event of the last quarter landed on them, where they sit against the band for their level, and, for anyone who joined this year, whether their first hour here was spent on them or on you. That is the entire retention half of this discipline on one line per person, and it needs no system of its own to hold it. The names take five minutes and the honest answers take twenty, and the twenty minutes is the point, because you'll write down at least one name you haven't thought carefully about since the day you hired them.

Members only · 4 principles + template + AI mentor

364 more frameworks are waiting.

The Hiring Operating Review is Day 1 of 365. One framework every morning for a year — across strategy, sales, negotiation, leadership, and more.

$1 /day

Billed $365/yr · cancel any time

365 frameworks

12 topics · templates · AI mentor

Reserve your place — $365/yr →

One payment a year · the whole library · no auto-renewal.

More Hiring frameworks

All 19 hiring frameworks →