I keep seeing candidates nail the arithmetic and still miss what interviewers are screening for: can you make a decision with imperfect data. When I’m coaching, I push a simple rhythm: open by stating the decision context (go/no‑go, invest, price, capacity). Declare what you’re including and excluding and why, tied to how the business actually makes money. Anchor one or two assumptions to something that would matter if you were operating this for real (willingness to pay, throughput, gross margin, distribution reach), then show the constraint that caps the upside. Run a quick sensitivity on the one assumption that moves the outcome, not all ten. Close with a recommendation and the next piece of information you’d validate tomorrow.
That typically reads as judgment rather than calculator work. For those who’ve sat on the other side: what phrases, guardrails, or quick checks signal “this person gets how businesses operate” during a sizing?
if you want “business judgment,” stop flexing decimal points. open with the decision and a kill‑rule: “if gross margin after variable costs < 25%, we pass.” cut vanity segments. cite a real constraint (capex, shelf space, physician throughput) and show how it caps revenue. do one triangulation, then decide. narrate trade‑offs, not spreadsheet theater. interviewer wants pruning, not polishing. and say what you’d validate next week if hired. simple, but somehow no one does it.
no one cares that your tam is 17.3b. pick a lane: “we’re testing a $50m beachhead with ~30% gross margin by year 2, or we walk.” lay out two paths and choose. call a risk explicitly (distribution or regulatory) and the gating datum you’d validate tomorrow. if you stack assumptions without choke points, it’s a fairy tale. keep it messy‑but‑decidable, not neat‑but‑useless. also, breathe. panic math reads like panic.
i started opening with: “we’re deciding go/no‑go for a $20m pilot.” then i add a rule: “if payback > 18 months, we pass.” interviewer actually nodded. tiny tweak, felt less wobbly. does that read as judgement to you?
i anchor one number to a public benchmark (arpu or utilization), then do a quick sensitivity on just that. feels more “operator” than student math. any better way to phrase it so it doesn’t sound rote?
small hack: i say what i’d test tomorrow (price ladder or conversion funnel step). it helps me sound decisive vs. speculative. is that overkill in a 10‑min sizing?
Two practical signals consistently land well with interviewers. First, frame the economic threshold before you calculate: for example, state the minimum gross margin or payback window that would make the opportunity viable for this business model. That immediately ties your math to decision criteria. Second, introduce a binding constraint early and quantify its effect: capacity limits, distribution coverage, regulatory throughput, or adoption friction. This shows you understand upper bounds, not just blue‑sky demand. If you have time, triangulate once from a different angle (bottom‑up units vs. top‑down spend share) and explain why the gap exists. Close with a recommendation and one specific piece of evidence you’d gather next. That arc—criteria, constraint, triangulation, decision—reads as judgment.
Love this! Framing the decision first is gold. Add one constraint, one sensitivity, then decide. You’ve got this—clear, confident, business-minded. Keep it crisp and you’ll stand out.
Signal judgment with a concise decision frame and one quantified constraint. Example: “Target a $30–50M beachhead if we can hit >30% gross margin and sub‑18 month payback.” Build the estimate using one anchor (e.g., 12M households, 10% serviceable, 15% adoption, $12 ARPU), then cap with capacity (2M monthly slots at 70% utilization). Run a ±20% sensitivity on adoption only and state the decision. Close with the next data point you’d validate (conversion from trial to paid). That sequence reads deliberate and credible.