In this article
Before you start
Why interviewers ask this
A manager's choices about people shape every project the team ships, so interviewers spend a whole round on them. The question families are repeatable: managing an underperformer, keeping a strong performer, resolving conflict, growing careers, hiring and owning a mistake, so you can prepare for them in advance. Many employers score answers with a structured interview, which means the same questions for every candidate and a rubric for each answer. Google describes using the same questions, an identical grading scale and rubrics that illustrate poor, borderline, solid and outstanding answers (Google re:Work, March 2026).
The people management round from first principles
- People management round
- A people management round is an interview in which a candidate for a manager role answers questions about hiring, feedback, performance, conflict and growth, usually with examples from teams they have led, and is scored on judgement and specifics.
- Question family
- A question family is a group of differently worded interview questions that test the same skill, such as handling an underperforming engineer.
In plain words
Preparing for the round is like packing for a trip with several climates. You do not pack one outfit per forecast. You pack a few pieces that combine. A story bank works the same way. It holds a few real stories, and each one is flexible enough to answer more than one question. The analogy breaks where honesty matters: a story must be yours, because follow-up questions check the detail. Said precisely, build a true story for each territory, shape each with a result and a change, and practise matching them to questions out loud.
Step 1: Learn the six territories
The round keeps returning to six territories, and knowing them tells you what to prepare. They are managing an underperformer, keeping a high performer, resolving conflict, growing someone's career, a hiring decision and a mistake that affected a person. The output is six empty slots to fill, one for each territory, so you can see what is still missing. Three territories have their own page: the low performer answer, growing engineers and conflict between two engineers. The one-on-ones and hard feedback page covers the hard conversations that run through all six. What you now know: what to prepare, not yet how.
Step 2: Pick one real story for each
A good story is a real event that you led, with a clear result and something you would do differently. The output is one chosen event per territory. Choose events that were hard, not flattering, because the interviewer probes the difficult parts. Describe people by role and keep names out. If a territory has no story, note it now; the trade-offs section covers the gap. What you now know: which six events you will tell.
Step 3: Shape each story
Shaping a story means cutting it to the situation, your actions, the result and what you changed afterwards. The input is the raw event; the output is a one-minute account. Spend most of the time on what you personally did, since the interviewer scores your judgement. For feedback moments, say what you observed and its effect, then ask for the other person's view, which is the situation-behavior-impact-intent (SBII) approach the Center for Creative Leadership (CCL) describes (CCL, August 2025). The situation, task, action, result guide shows the timing. What you now know: each story fits in about a minute.
Step 4: Map each story to the questions it can answer
Mapping a story means listing which question families it can answer, so one story serves several questions. The input is your six stories and a list of the question families. The output is your own grid, with stories down the side and families across the top. Mark where a story could honestly answer a family, and look for any empty column. How many marks each story earns depends on how you tag your own stories, so no count of stories is right for everyone. Fill any empty column with another story or a hypothetical answer. What you now know: where your gaps are before an interviewer finds them.
Step 5: Rehearse aloud and time it
Rehearsing means saying each story aloud, with the opening sentence ready and the length checked. The input is the shaped story; the output is a version you can say without notes. Say it to a person or record yourself, and trim any part that does not carry the action or the result. Practise the bridge sentences too: "the same story shows how I give feedback", so you can reuse a story without repeating it word for word. Practise one hypothetical answer for each territory as well. What you now know: you can start any answer in one clear sentence.
Step-by-step description
- Learn the six territories.
- Choose one real story for each.
- Cut each to situation, action, result, change.
- List the questions each story can answer.
- Say each story aloud and time it.
- The round keeps returning to six territories, and knowing them tells you what to prepare.
- Mapping a story means listing which question families it can answer, so one story serves several questions.
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
A reader cannot judge an answer without seeing what the levels sound like, so this section shows the rubric and then two answers to the same prompt. Everything here is an illustrative example written for this page. The rubric is not any employer's scoring sheet, the answers are invented third-person composites, and no real company or person is described.
| Dimension | Weak | Strong | Staff |
|---|---|---|---|
| Judgement | States a belief about people with no situation behind it. | Chooses a response to a real situation and says why. | Explains the alternatives rejected and the risk accepted. |
| Specifics | "We talked and it improved." | Names the gap, the plan, the dates and the signal watched. | Adds what the signal said at each review and what changed. |
| Ownership of action | Describes what the team or human resources (HR) did. | Says what the manager personally did, step by step. | Also says what the manager chose not to do, and why. |
| Result and change | No outcome, or an outcome with no evidence. | States the outcome and how it was known. | Names the change made to the team's process afterwards. |
| Follow-up readiness | Detail collapses under one probing question. | Holds up to two or three follow-ups on the same event. | Anticipates the follow-up and answers it unprompted. |
Both answers below respond to the same prompt, "Tell me about a time you dealt with an underperforming dev?", and the words that make the difference are in bold.
Weak answer (illustrative example). A candidate says: "I believe in giving people room to grow. I had a developer who was struggling, so we talked it through a few times and things got better over time. The team worked together and we improved delivery."
Strong answer (illustrative example). A second candidate says: "A developer's pull requests had been waiting on rework for three weeks. I first asked what was in the way and found they had never been told the review standard. I wrote the standard down with them, set a check every Friday for six weeks, and took one project off their plate so the plan was fair. By the fourth Friday reviews passed first time. Afterwards I added the standard to every new hire's first week."
Counting the actions and results an interviewer could probe gives 0 for the weak answer and 6 for the strong one.1 Reading the pair: the weak answer has a belief, no dates, no personal action and no result. The strong answer has a cause found, a written plan, a stated review rhythm, a decision about workload, a measurable outcome and a change to the process. Neither answer needs a dramatic event. The strong one is better because every sentence can be questioned and still stand, which is what the follow-up readiness row measures.
Text description of the diagram
- Counting the details an interviewer could probe gives 0 for the weak answer and 6 for the strong one.
- The strong answer has a cause found, a written plan, a stated review rhythm, a decision about workload, a measurable outcome and a change to the process.
Trade-offs and edge cases
The round can ask for a story or a hypothetical, and a good candidate prepares for both. The U.S. Office of Personnel Management (OPM) says a structured interview may ask about past behaviour, hypothetical situations or both (OPM). A past-behaviour question asks what you did and checks the detail of a real event. A hypothetical question asks what you would do and checks your reasoning and principles. That is why you prepare one story and one hypothetical answer for each territory.
Text description of the diagram
- A structured interview may ask about past behaviour, hypothetical situations or both.
- If you have no story for a territory, say so honestly and describe how you would act.
Three edge cases come up. First, you may have no story for a territory, such as letting someone go: say so honestly, then describe how you would act and who you would involve, as GitLab's handbook asks managers to involve its team-member-relations group early (GitLab Handbook, September 2025). Second, a story may involve confidential facts, so tell it by role and keep names and private details out. Third, company policy and local law differ, so describe your principle and say you would follow the employer's process.
How to answer it in an interview
The model answer below responds to the opening question of the round. It names the five jobs, gives the method and signals that examples follow. It is a method, not a past event, so you can say it truthfully and then attach a real example of your own.
How do you approach managing people?
I think of five jobs: setting clear expectations, giving specific feedback, growing each person, resolving conflict early and hiring on evidence. When someone's delivery slips, I find the cause first, agree a written plan with review dates, and protect the team's workload while the plan runs. I decide at the end and I say what I would change in how I set expectations. I can give you an example of that, if you like.
Spoken structure: say the five jobs, state the method in one or two sentences, then offer an example. At that point, tell one of your own real stories from the bank in the shape of the strong answer above. Keep the answer to about a minute and let the interviewer pick the next thread.
Common mistakes
- Philosophy without a story. (wrong) "I believe in empowering people." (right) "Here is a time I gave an engineer ownership of a design and what happened."
- A story with no result. (wrong) "We talked it through." (right) State the outcome and how you know.
- Taking credit for the team. (wrong) "We improved delivery." (right) Say what you personally did.
- Naming the person. (wrong) A full name and private details. (right) Describe by role and keep details private.
- One story for everything. (wrong) The same example for six questions. (right) Use a different story for each territory, each bent to its question.
Follow-up questions
Have you ever had to let someone go?
Answer yes or no plainly. If yes, describe the documented feedback, the plan, who you involved and how you treated the person; if no, say so and describe how you would handle it. The low performer answer walks through the whole sequence.
What if you have never let someone go?
Say so plainly, then describe how you would approach it: documented feedback, a clear plan, involving your human resources team early and treating the person with dignity. Interviewers value honesty over an invented story, and the low performer page shows how to build that hypothetical answer.
What is constructive feedback you've gotten?
Pick real feedback that you acted on, say who gave it, what you changed and what was different afterwards. Choose something that a manager could actually fix, not a disguised strength. The one-on-ones and hard feedback page covers giving and receiving it.
What was the problem, and how did you work with them?
State the problem as an observable behaviour and its effect, then describe the conversation, the plan and the result in your own actions. Keep the person described by role. The conflict between two engineers page has a full example of the method.
What are engineering manager behavioral interview questions?
Engineering manager behavioral questions ask about a past situation, such as an underperformer, a conflict, a hiring call or a mistake, and a people management round is largely made of them. A behavioral round covers any role's competencies and this round focuses on managing people, so see the behavioral interview guide and how the interview loop fits together.
How do you talk about team-wide health, not only individuals?
Name a small set of signals for the whole team, such as how reviews flow, how often people raise problems and who is carrying the load. Then say what you do when one signal moves, for example a short team discussion followed by one change. The team health page covers the full answer.
- The round tests judgement and specifics through stories about people.
- Prepare a real story for each of the six territories.
- Shape each one: situation, your actions, result, what you changed.
- Map each story to the question families it can honestly answer and fill any gap.
- If you lack a story, say so and describe how you would act.
Sources
- Structured Interviews — U.S. Office of Personnel Management (undated page; date = read date), 24 Oct 2026
- A guide to structured interviewing for better hiring practices — Google re:Work (updated March 2026), 1 Mar 2026
- Underperformance, The GitLab Handbook — GitLab (content last updated 2025-09-04), 4 Sep 2025
- Closing the gap between intent vs. impact: the SBII feedback model — Center for Creative Leadership, 19 Aug 2025
