In this article
Before you start
Why interviewers ask this
Stakeholder questions test whether you can keep partners aligned without becoming a messenger or a wall. A candidate who says "I keep everyone updated" has not said who gets which update, who decides when partners disagree, or what happens when you must say no. The question comes in several shapes: how you manage stakeholders and protect the team, how you align engineering and product when they disagree, how you say no and keep the relationship, and a time you lost a stakeholder's trust and repaired it.
A weak answer describes meetings. A strong answer shows a method: who the stakeholders are, what each one optimises for, who decides, and how bad news travels. It includes one story where the relationship was tested and what changed afterwards. Planning questions such as how do you build a roadmap and prioritise are the usual neighbour of this one.
Stakeholder management from first principles
- Stakeholder
- A stakeholder is a person or group who depends on your team's output, funds it, constrains it or is affected by its decisions, and whose agreement or information you therefore need to deliver the work.
- Power-interest grid
- A power-interest grid is a two-by-two sort of stakeholders by how much they can influence your work and how much they care about it, which tells you how much attention and which kind each one needs.
- Decision owner
- A decision owner is the single named person who makes the final call on a given decision, after hearing the others, so that disagreement ends in a decision instead of a stalemate.
In plain words
Managing stakeholders is like running a kitchen pass. Orders come in from tables with different needs, the cook knows what is possible, and the person at the pass tells each table honestly when its dish will arrive. The kitchen stays calm because nothing is promised that cannot be cooked. The analogy breaks on power. A table cannot overrule the chef, but a stakeholder with budget or a senior title can, so you must agree who decides.
Said precisely, the work has five steps. List the stakeholders and what each optimises for. Map power and interest to set a rhythm. Agree who decides. Share bad news early with a trade-off attached. Repair trust fast when it breaks. Google's research on its best managers found that "communicate effectively", "implement a clear vision and strategy" and "collaborate across Google" are among its ten manager behaviours (Google re:Work, August 2025).
Step 1: List the stakeholders and what each optimises for
The first step is a written list of everyone who depends on, funds, constrains or is affected by your team, with what each one wants. The input is your current projects; the output is a table of names, the outcome each person is judged on, and the worst thing that could happen to them. A product manager is judged on a launch, a finance partner on a budget, a support lead on ticket volume, a legal reviewer on risk.
Writing the objective next to each name turns a vague "alignment" problem into a set of specific trades. It also shows who is missing: the person who learns about your change from a customer is a stakeholder you did not list. Include your own team, which depends on you for context and cover. What you now know: who matters and what each one needs.
Step 2: Map power and interest and set a rhythm
The second step sorts the list by power and interest and gives each group a different rhythm. The input is the list; the output is four groups. Score each person from 1 to 5 on how much they can change your plan and how much they care; a score of 3 or more counts as high. High power and high interest means manage closely; high power and low interest means keep satisfied; low power and high interest means keep informed; low on both means monitor.
In the worked example below, eight stakeholders sort into 4, 2, 1 and 1. Manage closely gets 60 minutes a week, keep satisfied 15, keep informed 10 and monitor none, which totals 280 minutes a week or about 61 hours over a 13-week quarter.1 What you now know: how your time is divided and why.
Step 3: Agree one decision owner who decides
The third step names a decision owner for each recurring decision before a conflict, so disagreement ends in a decision. The input is the decisions that repeat, such as scope for a release; the output is a written owner per decision. Atlassian's Team Playbook describes DACI, which stands for Driver, Approver, Contributors and Informed, and stresses that there is exactly one Approver (Atlassian, Team Playbook, read October 2026).
Netflix gives the same role to an informed captain who decides after "farming for dissent", because decision-making by committee "tends to slow companies down and undermine accountability" (Netflix culture page, read October 2026). Persuasion ends a dispute only when one side changes its mind; a named owner ends it even when neither does. Amazon's 2016 letter asks for "disagree and commit" when there is conviction but no consensus (Amazon, 2016 letter to shareholders). When engineering and product disagree on priority, the question is not who is right but who decides. What you now know: who closes a disagreement.
Step 4: Protect the team by sharing bad news early
The fourth step is to tell stakeholders about a slip, a risk or a no as soon as you know it, with the trade-off attached. The input is a changed fact, such as a project that will miss its date; the output is a short message with the new date, the cause, the options and your recommendation. Protecting the team means absorbing noise and giving it context. It does not mean hiding information from it or from partners.
A useful message has four parts: what changed, what it affects, the options with their costs, and what you recommend. The same shape works for a no: name what the request would displace, in the requester's units such as launch date or revenue risk, and offer the order you would use. Partners forgive a late project far more often than a surprise. What you now know: how bad news travels without damage.
Step 5: Repair trust quickly when it breaks
The fifth step is a short, honest repair when a stakeholder loses trust, because trust lost to a surprise rarely returns by itself. The input is the broken commitment; the output is an acknowledgement, a cause, a fix and a new habit. Say plainly what went wrong, without defending it. Explain the cause in one or two sentences and say what you changed in how you work, such as a weekly risk note or an earlier check-in.
Then keep the new habit visible for a few weeks so the stakeholder can see it working. An interview story about this question is strongest when it includes the stakeholder's own later words or actions, such as asking you to run the next launch. What you now know: the relationship can be rebuilt, and how to show it.
Step-by-step description
- Who depends on you, and what each is judged on.
- Power and interest: 280 minutes a week.
- One decision owner per decision.
- Bad news early, with options.
- Acknowledge, fix, show the new habit.
- The mapped plan costs 280 minutes a week.
- DACI names one Approver per decision.
Knowing it is half the job. Saying it is the other half.
Interview Clozo is an AI interview copilot for real interview rounds: it hears the question and puts a strong answer on your screen, and you say it in your own words. See the features or how it works.
Start Free Trial →Worked example
The example uses eight made-up stakeholders to show how the grid turns into a weekly plan. Scores run from 1 to 5, the split is 3, and the minutes per week by group are illustrative. Every number below comes from a short script with saved output.1
| Stakeholder | Power | Interest | Group | Minutes a week |
|---|---|---|---|---|
| Product manager | 5 | 5 | Manage closely | 60 |
| Engineering director | 5 | 4 | Manage closely | 60 |
| Design lead | 3 | 5 | Manage closely | 60 |
| Data science lead | 3 | 4 | Manage closely | 60 |
| Finance partner | 4 | 2 | Keep satisfied | 15 |
| Legal reviewer | 4 | 1 | Keep satisfied | 15 |
| Support lead | 2 | 4 | Keep informed | 10 |
| Sales engineer | 1 | 2 | Monitor | 0 |
Giving every stakeholder 60 minutes a week would cost 480 minutes, which is 104 hours over the quarter, or 43 more hours than the mapped plan.1
Text description of the diagram
- The mapped plan costs 280 minutes a week.
- Giving every stakeholder 60 minutes a week would cost 480 minutes.
Trade-offs and edge cases
The grid is a starting point, not a verdict, and several cases break it.
- A low-interest, high-power stakeholder who can veto. Legal and finance score high on power and low on interest, and a surprise from them is expensive. Keep them satisfied with a short, early note, not a silence.
- Interest that changes. A support lead's interest jumps when your launch will double tickets. Re-score at each milestone.
- Conflicting stakeholders. When two high-power partners want opposite things, do not average them; surface the conflict to the decision owner with the trade-off.
- Protecting the team versus transparency. Absorb noise, such as shifting requests, but share facts that change the team's work. A team that hears a risk late loses trust in you.
- Too many stakeholders. If the list passes ten, the problem may be unclear ownership of your team's scope, not a communication problem.
| Group | Action | Minutes a week | People in the example |
|---|---|---|---|
| Manage closely | Decide together | 60 each | 4 |
| Keep satisfied | Early, short notes | 15 each | 2 |
| Keep informed | Regular updates | 10 each | 1 |
| Monitor | Watch for change | 0 | 1 |
Text description of the diagram
- A score of 3 or more counts as high.
- Four stakeholders are high on both power and interest and get 60 minutes a week each.
How to answer it in an interview
How do you manage stakeholders, protect your team and stay aligned?
I start by listing who depends on my team and what each person is judged on, then sort them by power and interest so my time follows both. For recurring decisions I agree one decision owner in advance, so disagreement ends in a call. When something slips, I tell partners early with the cause, the options and my recommendation, and I tell my team the same facts. Once a launch date slipped and a partner was surprised. I apologised, started a weekly risk note, and they later asked me to run the next launch.
The spoken structure is: who and what they optimise for, how your time follows power, who decides, how bad news travels, and one repair story.
Common mistakes
- Shielding the team from information. (wrong) "I protect my team from the noise, so I don't share the risks." (right) Absorb the noise, share the facts that change their work.
- Updating everyone equally. (wrong) "I send the same weekly note to all twenty people." (right) Different groups get different actions and rhythms.
- No decision owner. (wrong) "We discuss until we agree." (right) Name who decides, after dissent is heard.
- Late bad news. (wrong) "I waited until I had a full plan." (right) Share the slip with options as soon as you know.
- A repair story with no change. (wrong) "I apologised and we moved on." (right) Say what you changed in how you work and what the partner did afterwards.
Follow-up questions
How do you align stakeholders when engineering and product disagree on priority?
Put both positions and the data in front of the agreed decision owner, with the trade-off in shared units such as launch date and customer impact. Then commit to the decision, as in the disagree-and-commit idea, and review it at an agreed date. For ordering questions, see how to build a roadmap and prioritise.
Tell me about a time when you lost the trust of a stakeholder. How did you repair the relationship?
Name the broken commitment, the cause in one sentence, what you did within a week, and the new habit. End with something the stakeholder did later that shows trust returned.
How do you communicate that a project will slip to stakeholders?
Tell them as soon as you know, with the new date, the cause, the options and your recommendation. For a longer example of changing course, see when priorities change mid-quarter.
How do you say no to a stakeholder and keep the relationship?
Show what the request would displace, in their units, and offer an alternative order or date. Ask which committed item they would drop; a no with a trade-off is an offer to decide together.
- List stakeholders with what each is judged on, and who is missing.
- Sort by power and interest; the example's plan costs 280 minutes a week.
- Agree one decision owner before a conflict, so disagreement ends in a call.
- Protect the team by absorbing noise and sharing facts, early.
- Repair trust with an acknowledgement, a cause, a fix and a visible habit.
Sources
- DACI: a decision-making framework, Team Playbook — Atlassian (page undated; read 2026-10-24), 24 Oct 2026
- 2016 Letter to Shareholders — Amazon (Jeff Bezos), 17 Apr 2017
- Netflix Culture — Netflix (page undated; read 2026-10-24), 24 Oct 2026
- Following the data: the research behind great managers — Google re:Work (August 2025), 1 Aug 2025
