the read you are after
You are not trying to find out if a candidate can perform product thinking in a room. You already know they can — they made it past your resume screen. What you can’t yet see is whether, at 9pm on a Friday when the data is split and the exec is confident and the team is tired, they make the call you’d trust. That’s the read.
The pressure-test version of this question has a specific shape: does this person hold a decision when the room pushes back, or do they reach for “let’s get more data” as a deferral mechanism? Do they reframe a problem that’s been posed wrong before the room names it? Do they own a miss cleanly, or do they rebuild it as a lesson about external conditions? None of these show up in a case study. They show up when there’s actual pressure in the room and something at stake — even simulated stakes inside a well-designed round.
The loop’s job is not six separate reads on general product instinct. It’s to assemble a fingerprint — two competencies per round, six rounds, panel calibrated on the L1–L4 behavioral anchors before interview day. Done right, you end the loop knowing the candidate’s shape across the framework, not just their average impression. A spiky L4 on signal and a L1 on hold is a fundamentally different person than a flat L3 across twelve — and those two shapes have different implications for your team depending on where your current gap is.
The hire is a shape match against what the team already has, not a points-total against the candidate alone.
the three competencies that decide most hires
Hold — Resolve skill
Why it’s diagnostic: hold is the competency you can’t prep for, can’t mimic with frameworks, and can’t fake under realistic pressure. Every other competency can be performed. Hold — actually staying in a position when a senior person pushes back without hostility, just authority — either happens or it doesn’t.
L4 signal: they hold the call, name the specific point of disagreement, and address the pushback with evidence rather than re-explaining. They don’t get louder, they don’t get softer — they stay at the same register and keep pointing at the data.
The cargo-cult tell: they “hold” by deferring — “you may be right, let me take that back and think on it more” sounds like openness; it’s actually avoidance. Real hold doesn’t retreat to process. Watch for the candidate who keeps escalating the ask for data rather than committing to the call on the data they have.
Reframe — Acuity skill
Why it’s diagnostic: give a candidate a solution-shaped problem and watch what they do with it. Most PMs solve the problem as posed. A small fraction notice — within the first few minutes, before anyone else names it — that the question is wrong. That upstream move, done without ceremony, is the Reframe signal. It separates candidates who are sharp from candidates who are merely fast.
L4 signal: they name the reframe out loud and explain why the posed version would lead to the wrong answer — not as a critique of the question, but as a natural part of picking up the problem. “Before I answer that, I want to make sure we’re asking the right thing here…”
The cargo-cult tell: they reframe loudly as a performance — “actually, I’d challenge the premise of this question” — without a better frame to offer. Real Reframe produces a sharper question, not a better meta-commentary on the original one.
Miss — Resolve skill
Why it’s diagnostic: every candidate has a miss. The question is what they do with it. Miss is the cleanest read you have on whether someone builds an honest relationship with evidence. Ask them to walk through their worst call — not a “growth opportunity,” the actual decision that failed. The gap between levels is clear and fast: deflection, normalization, ownership, and learning are four distinct behaviors that surface in about eight minutes.
L4 signal: they name the call, name what the evidence actually said at the time, own what they misread, and tell you what it changed about how they make that category of decision now. The past tense is clean — no residual defensiveness, no narrative about why conditions were unusual.
The cargo-cult tell: the “growth opportunity” frame — phrasing a miss as a learning that made them stronger, without ever acknowledging what the decision actually cost. It’s not dishonesty; it’s pattern. L1 Miss isn’t lying, it’s the habit of building stories that don’t include a clean reckoning.
what to do this week
Before your next loop opens: write down your current team’s MARK fingerprint — even an informal read against the twelve competencies. Then write down the two or three competencies that represent the sharpest gap. Send that gap description to every panelist before the loop, not during the debrief. If you can’t write the fingerprint, you’re not ready to design the loop — and the loop will default to selecting on articulation, which is the wrong screen.
read this next
Read the full playbook: MARK for hiring — the 6-round loop design, the round-by-round competency targeting, the Brief as pre-screen, and the scoring and debrief protocol.
avoid these traps
Calibrating after the loop instead of before. If panelists have different mental models of what L3 hold looks like, the debrief is a negotiation, not a read. Walk through behavioral anchors before interview day — thirty minutes buys you two hours of debrief alignment.
Hiring on Brief score alone. The Brief is written under zero pressure. The Resolve skill — hold, power, miss — is invisible in written work. A strong Brief advances a candidate into the loop; it does not replace the loop.
Averaging the fingerprint. A 2.7 mean tells you almost nothing. A spiky L4 on signal with a L1 on hold is a specific risk profile. A flat L2 across twelve is a different one. Read the shape. Match it to the gap.
citation
PL Standard v3.1 · MARK for hiring panels