The Highest Best Use Interview
Ban hypotheticals, ask for lived history, and finish by asking what a great day looks like.
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 4
- Confidence
- 91%
Conrad's interview design has two rules. First, never ask candidates to speculate about a problem from your current business — it creates a knowledge imbalance that breaks the collaborative nature of a good interview. Almost every question should instead start with 'tell me about a time in your career when...'. Second, close every interview by asking the candidate to describe a great day at work, which reveals what they gravitate toward when left to their own devices — the key to placing them at their highest and best use.
Origin
Tom Conrad's own interviewing practice, developed across hiring at Pandora, Snap, Quibi, and Zero.
Core principles
- 01A hypothetical drawn from your live business tests the candidate on facts only you possess — it measures nothing but confidence.
- 02Past behaviour, specifically recounted, is the highest-signal evidence available in an interview.
- 03A great interview is collaborative and symmetric, not a trap.
- 04What someone naturally does when unsupervised is the thing they'll do more of tomorrow, because it rewards them.
- 05Hiring isn't finished at yes — the job is to find each person's highest and best use.
How to run it
- 1
Delete every question pulled from your live problems
Audit your question bank and remove anything of the form 'how would you solve this thing I'm working through right now?' Those questions force the candidate to speculate about a business only you understand.
Pro tip The tell that you've written one of these: you would be genuinely interested in the answer for your own roadmap.
Watch out These questions feel efficient (free consulting, apparent signal) but they create an imbalance that makes the candidate perform rather than think.
- 2
Reframe every probe as 'tell me about a time in your career when...'
Convert each competency you want to probe into a request for a real episode from the candidate's history relevant to that topic. Opening with this phrase also sets the expectation and gives permission to talk about things actually done.
Pro tip Use the same stem consistently so the candidate calibrates fast and stops guessing what you want.
- 3
Close with the great-day question
Near the end, ask: imagine you had a really great day at work — tell me what it was that you did on that day. Listen for what they choose unprompted, because that is the naturally rewarding activity they will seek out again.
Pro tip The answer is often more diagnostic than the whole rest of the interview, because it exposes the internal reward mechanism rather than the rehearsed narrative.
Watch out Do not lead the answer by describing the role's ideal day first.
- 4
Place the hire at their highest best use
Use the great-day answer to decide not just whether to hire, but where to put them. The natural reward mechanism tells you which shape of work they will compound in.
Pro tip Conrad's own career is the argument for this: forced into a narrow lane at Berkeley Systems, he finally became a real engineer instead of remaining 'six miles wide and a centimetre deep.'
Watch out A great candidate placed against their natural reward mechanism will underperform and you will blame the hire.
In the wild
Asked about hobbies, Conrad realizes he doesn't really have any: what he does for fun on a weekend is go deep on some piece of interesting software, like teaching himself After Effects. He loves writing software, building software, and using software. This is the same signal the great-day question is designed to extract from a candidate.
→ That alignment — what he loves, what he's good at, what the world needs, and what he can be paid for all overlapping — is what he credits for a 30-year career without burning out.
At Apple, Conrad was rewarded for contributing across functions — engineering, design, product marketing — and left after four years feeling like a 'professional gadfly,' six miles wide and a centimetre deep, having shipped little. At Berkeley Systems, running the You Don't Know Jack engineering team, his job was explicitly to ship high-quality software on time; insightful feedback on gameplay or packaging would benefit his career not at all.
→ Being forced into a lane is when Conrad says he actually became a software engineer — his college education and four years at Apple had both been insufficient. His takeaway: it really does pay to get good at something rather than be a generalist, and cultures with swim lanes let people be exceptional in theirs.
Common mistakes
Using the interview as free consulting
Asking candidates how they'd solve your current problem makes them speculate about a business you know intimately and they know nothing about. It measures bravado and creates an interviewer/interviewee imbalance.
Stopping at the hire/no-hire decision
The point of the great-day question is placement, not just selection. Failing to act on what you learn wastes the highest-signal answer you got.
Rewarding generalist breadth by default
Cultures that reward broad cross-functional contribution can produce people who are wide and shallow and ship little. Sometimes the kindest thing is a lane.
Is it for you?
Best for
Hiring managers, product and engineering leaders building a team from scratch, and anyone whose interview loop currently consists of hypothetical case questions from the live roadmap.
Not ideal for
Roles where the job genuinely is speculative strategy on your specific business (e.g. a paid trial project), or high-volume entry-level hiring where candidates have no career history to recount.
From the transcript
“because just causing the other person to like speculate often times like about your own business”
“almost every interview question should start with tell me about a time in your career when”
“imagine that you had like a really great day at work tell me about what like what was it that you did on that day”
“it's really important to find the highest best use of everyone”
From the episode
Billion dollar failures, and billion dollar success
Tom Conrad (Quibi, Pandora, Pets.com, Snap, Zero)