ChannelPulse
The story inventory, built by coverage
Most people prepare stories by collecting impressive projects, then discover in the room that four of them evidence the same two signals. This lesson builds the inventory as a coverage matrix, decomposes one project into several genuine stories, and settles when reusing a story is fine and when it costs you.
What you'll be able to do
- Build a story inventory as a coverage matrix over the signals, and find the holes
- Decompose a single project into several distinct stories without inventing anything
- Decide when to reuse a story across questions and when variety is worth the weaker answer
- Recognise a story carrying too many signals, and why that is a risk rather than efficiency
Before this: star-without-sounding-like-star
The standard advice is to prepare five or six strong stories, and it is not wrong, but it's the wrong unit. Stories are not the thing being collected — signals are. Six stories that all evidence perseverance and communication is a thin form, however impressive the projects were.
So build the inventory the way the scorecard reads it: as a matrix.
Useful here not as answers to copy but as a coverage test. Each chapter is one real question — play them in turn and ask which of your stories you would reach for. The questions you cannot answer twice are the holes the matrix below is meant to find.
Jump to the part you need
The coverage matrix
Down the side, the eight focus areas most tech behavioural rubrics collect signal on. Across the top, your stories. A mark goes in a cell only when the story contains a specific moment you could be pushed on for that area — not when it merely touches it.
Here is a plausible starting inventory:
| Focus area | Payments migration | On-call outage | Disagreed with staff eng | Missed the Q3 date | Covered by |
|---|---|---|---|---|---|
| Motivation | 0 | ||||
| Proactivity | ● | 1 | |||
| Ambiguity | ● | 1 | |||
| Perseverance | ● | ● | 2 | ||
| Conflict | ● | ● | 2 | ||
| Empathy | ● | 1 | |||
| Growth | ● | 1 | |||
| Communication | ● | ● | 2 | ||
| Areas carried | 3 | 2 | 3 | 2 | 7 of 8 |
Four stories, seven of eight areas covered, and one specific hole: motivation. That is a useful place to be, because the fix is one story rather than a rewrite — and notice that motivation is the signal people forget precisely because it feels like the softest one. "What are you proud of and why" sounds like small talk. It is a scored box.
The column of counts is the other output to take seriously. Four areas — proactivity, ambiguity, empathy, growth — rest on exactly one story each. If the interviewer never asks the question that surfaces the disagreement story, empathy goes uncollected — not because you lack it, but because your only evidence for it was contingent on a question being asked. Coverage of two is meaningfully safer than coverage of one.
Now read the bottom row instead. A story carrying four or more areas looks efficient and is actually concentration risk: if the interviewer pushes on one detail and it wobbles, everything resting on that story wobbles at once. In this inventory the maximum load is three, which is a reasonable ceiling.
One project, several real stories
The common objection at this point is that you don't have enough material — one substantial project, maybe two. Usually that's a framing problem rather than an experience problem.
A project is not one event. It's a sequence of decisions, and the moments inside it are genuinely different stories with different signals:
| Moment in the project | The story it becomes | Signals it evidences |
|---|---|---|
| Before it was a project — you noticed the problem | "I saw the nightly job was going to hit the traffic ramp within two quarters" | motivation, proactivity |
| The scoping, when nobody agreed what it was | "The first two weeks were spent turning 'fix payments' into three specific workstreams" | ambiguity |
| The disagreement about approach | "The staff engineer wanted to parallelise; I thought locks were the constraint" | conflict, empathy |
| The part that went wrong | "My cutover plan assumed idempotent retries. It didn't hold, and we double-charged 40 accounts" | growth, perseverance |
| The grind | "Weeks four through nine were blocker after blocker" | perseverance |
| The handoff | "Making it someone else's to run, including the runbook I'd have wanted" | communication, empathy |
That is one project and six answers, none of them invented. You're not manufacturing experience; you're spotting the moments in the experience that show how you think — which are the moments that usually get compressed into "and then we shipped it."
Reusing a story: the honest answer
Candidates worry about repetition more than interviewers do. The rule practitioners converge on is blunt: the best answer to the question wins, even if it reuses the project. An interviewer would rather hear one detailed, compelling example that lands the signal than three mediocre ones from different projects. If your migration is genuinely your best conflict story, use it for the conflict question — you already used it for ambiguity, and that is fine.
The refinement is about context:
- Within a single round, repetition matters least. You're getting five or six questions from one person; reusing a project two or three times with clearly different moments reads as depth. Reusing the same moment twice reads as having one story.
- Across a loop, it matters more. If four interviewers each hear about the payments migration, the debrief has one data point on you rather than four, and someone will say so out loud. Vary the project across interviewers when the quality is comparable — and only then.
- When quality is not comparable, take the better answer and accept the repetition. A weaker story told for the sake of variety is a worse trade than a repeated project, because it fails the question you were actually asked.
Building yours
Concretely, an hour's work:
- List every project from the last two or three years, including the ones you don't think of as impressive. The unglamorous ones often hold the best conflict and ambiguity moments.
- Decompose the two biggest into moments using the table above — noticing, scoping, disagreement, failure, grind, handoff.
- Draw the matrix with those moments as columns, and be strict about which cells you mark.
- Read the zeroes, and write to them. One story per hole, and prefer two-deep coverage for conflict and growth, which are the areas most reliably asked.
- Read the bottom row. Split anything carrying four or more areas into two moments.
What's next
You now have an inventory that covers the form. The next question is whether the stories in it are big enough — because the same story, told identically, is a hire at one level and a down-level at the next, and that has nothing to do with how well you tell it.