Behavioral & STAR · The STAR Method · Lesson 2 of 3
Quantifying the Result
An unquantified result is an anecdote; a quantified result is evidence. The number is what the interviewer writes down and repeats in the debrief. If you take one thing from this track: every story needs a number.
- Prefer hard metrics: %, $, time saved, latency, users, tickets, revenue, error rate.
- No metric handy? Estimate honestly and say so: 'roughly a 30% drop in support tickets — I sized it from the queue before and after.'
- Use a proxy when direct numbers don't exist: adoption, repeat usage, a promotion, a renewal, positive stakeholder feedback you can name.
- Close the loop: tie the number back to why it mattered to the business, not just that it moved.
Weak: "We improved the onboarding flow and it worked well." Strong: "I cut onboarding from 6 steps to 3, which lifted activation from 48% to 71% in one quarter — about 1,200 more activated users a month."
Vague = forgettable'Significantly', 'a lot', 'much faster', 'the team was happy' — these evaporate in the debrief. If you can't measure it, at least scope it (before/after, how many, how long).
◆ Lock it in
- A number turns an anecdote into evidence the interviewer records.
- Use hard metrics; if none, estimate honestly or use a proxy.
- Tie the result to why it mattered to the business.
Feynman drill — say it out loudTake a vague result ('it went well') and rewrite it with an honest, scoped number.
Step 1 rate your confidence · Step 2 pick your answer
You genuinely don't have a precise metric for your result. Best move?
Step 1 — how sure are you?