Behavioural interviews · Module 08 of 12 · Level 2: The core questions

Ambiguity, priorities and trade-offs

"Tell me about a time the requirements were unclear." "…a time you had competing priorities." These questions test judgement: how you create clarity, how fast you move, and whether you can say plainly what you chose not to do.

1. The idea

Sailors crossing in fog don't wait for it to lift. They take a bearing, go slowly, sound the horn, check the depth often and correct course as soon as the coast appears. Ambiguity questions ask whether you can make progress in fog without either freezing or charging ahead blind.

Two related families:

Ambiguity: "A time the requirements were unclear", "a decision with incomplete data", "a project with no clear owner or direction".
Priorities and trade-offs: "Competing priorities", "a time you had to say no", "a trade-off between speed and quality".

What strong answers show:

You reduced the fog first. You asked the questions that mattered, found who could answer them, and wrote down what was still unknown.
You made your assumptions explicit and shared them, so others could correct them.
You matched the care to the risk. Decisions that are easy to undo can be made fast; ones that are hard to undo deserve more data. Amazon's principles put it as "Many decisions and actions are reversible and do not need extensive study."
You took a small step and learned from it: a prototype, a trial, a reversible first release.
You named the trade-off. What you chose, what you gave up, and who you told.

2. How it works

A vague request ("make onboarding better") turned into a decision. Watch the unknowns shrink at each step.

3. Try it yourself

Priorities under pressure

Five pieces of work, one engineer, two weeks. Set how much each matters, what it costs, and how soon it's needed, and tick the ones that would be hard to undo. The ranking explains itself.

4. What interviewers look for

Ambiguity is scored on how you created clarity and how deliberately you chose. Trade-off stories are scored on whether you named what you gave up.

Signal	Weak	Good	Strong
Clarity	Waited for someone to define it	Asked some questions	Found the right people, wrote down assumptions, and shared them
Pace	Froze, or charged ahead blind	Made a decision	Fast on reversible steps, careful on irreversible ones
Trade-off	"We did everything"	Explained the choice	Named what was given up, and told the people affected
Learning loop	No feedback	Checked the result	Small first step, measured, then adjusted

5. A model answer

"Tell me about a time you had to work with unclear requirements." An invented answer, shown part by part. Read it for the shape, then build your own from a true story.

SituationOur VP asked my team to "make onboarding better" for a B2B analytics product, with no metric and no spec. New customers were churning, but nobody knew why.

TaskI was asked to lead the work, with one designer and six weeks.

ActionFirst I turned the ask into a question I could measure: what share of new accounts create a first dashboard within a week? It was 31%. I wrote down my three biggest assumptions and checked them: I watched recordings of 20 sign-ups and called five churned customers. Two assumptions were wrong: the problem wasn't the tutorial, it was connecting a data source, where 60% of people stopped. I proposed a sample-data option, which we could ship in a week and remove if it didn't help, instead of rebuilding the connector flow, which would take a quarter. I told the VP the trade-off in writing: the connector rebuild would wait a quarter.

ResultFirst-dashboard rate rose from 31% to 52% in a month, and three-month churn in the next cohort dropped by about a quarter. We used what we learned to scope the connector rebuild for the next quarter.

LearningI'd have watched the recordings before the first planning meeting, not after: it would have saved a week of debating the tutorial.

Why it scoresThe candidate turns a vague ask into a metric, tests assumptions with real users, chooses a reversible step over a big irreversible one, and states the trade-off to the person who made the request.

6. Remember it

Memory hook: "Shrink the fog, then take a reversible step"Turn unknowns into questions, write assumptions down, and move first where it's easy to undo.

How this hook works: what “Shrink the fog, then take a reversible step” means

What the slogan means, phrase by phrase

“Shrink the fog”: Ask the questions that matter, measure what you can, and write down what's still unknown.
“then”: Clarify first, but not forever: a time box on the research.
“take a reversible step”: A small change you can undo, which teaches you more than another week of debate. Save the big, hard-to-undo moves for when the fog is thinner.

Why this picture fits: Sailing in fog: take a bearing, go slowly, check the depth, correct course. Waiting for clear skies means never leaving port; full speed means hitting the rocks.

The story → the concept

Sailing in fog	Working with ambiguity
Taking a bearing	Turning a vague ask into a measurable goal
Sounding the depth	Checking assumptions with data and users
Going slowly	A small, reversible first step
Correcting course	Adjusting after the first results
Telling the crew the route	Sharing the trade-off

Watch it on an example

"Make onboarding better" becomes "31% create a dashboard in week one".
Recordings show 60% stop at connecting data: an assumption corrected.
Ship a removable sample-data option in a week, not a quarter-long rebuild.

The trap it saves you from

Either waiting for perfect requirements, or building the big thing based on a guess.

It still works when…

The ambiguity is about people, not data. The same moves apply: ask, write down assumptions, try a small step, check.

7. Common mistakes

Waiting for clarity"I asked my manager for requirements and waited" shows dependence, not judgement.

Charging aheadBuilding the big solution on an untested guess. Show the questions you asked first.

No trade-off named"We managed to do everything" is rarely true and skips the judgement interviewers are scoring.

Every decision treated the sameReversible steps deserve speed; irreversible ones deserve care. Say which kind yours was.

Saying no badlyPriority stories should show how you told people what wouldn't happen, and what you offered instead.

8. Practice exercise (25 minutes)

Pick an unclear project (5 min). One where you started without a clear goal or spec.
Write the fog (5 min). List the three biggest unknowns at the start, and how each got answered.
Name the trade-off (10 min). What did you choose not to do? Who did you tell, and how?
Say it (5 min). Two minutes, recorded. Did you say whether the first step was reversible?

9. Self-check quiz

A decision is easy to undo. How much study does it deserve?

Reversible decisions are cheap to get wrong; speed matters more.

What's the best first move with a vague request?

Shrink the fog before committing.

Which sentence names a trade-off?

It says what was chosen and what was given up.

Why write assumptions down?

Written assumptions get challenged early, when it's cheap.

10. Go deeper

Resource	Time	Why
Bias for Action; Are Right, A Lot (Amazon)	3 min	Reversible decisions and calculated risk, in a company's own words.
Trade-off analysis (Software architecture, Module 06)	30 min	A structured way to compare options, useful for the story behind your choice.
Module 09: Company values	Next	Map your stories to what each company hires for.