What a structured interview question actually looks like in practice, role by role, with examples of strong and weak answers so you know what you're evaluating.
What Makes a Question "Structured"
Most interviews are conversations. The interviewer asks what comes to mind, follows whatever thread feels interesting, and ends up with a strong impression of some candidates and a vague one of others. The problem: impressions correlate more strongly with likability than with job performance.
Structured interviews fix this by using the same questions, in the same order, for every candidate. That standardisation lets you compare apples to apples. Research consistently shows structured interviews predict job performance roughly twice as well as unstructured ones. For a startup making a hire that could define the next 18 months, that difference matters.
Below are 30 structured interview question examples organized by role, with sample answers showing what a strong response looks like versus a weak one. Use them as-is or adapt them to your context.
Format used throughout this post
All questions use behavioral format ("tell me about a time...") because evidence from real past situations predicts future performance far better than hypotheticals ("what would you do if..."). When a candidate can't recall a specific example, that's itself a data point.
Question 1: Technical ownership
"Tell me about the last significant technical decision you made that you owned end-to-end. Walk me through the problem, your approach, and what happened."
What you're evaluating: Whether they take genuine ownership of technical decisions or wait for direction. Look for a clear problem statement, explicit reasoning for the approach chosen, and a honest account of the outcome including what they would change.
Weak answer
"We were building a new API layer and I wrote most of the code. It went well and the team was happy with it."Strong answer
"We had a latency issue on our data ingestion pipeline — jobs that should run in under 2 seconds were hitting 8 to 12 seconds at peak load. I diagnosed it as a database lock contention issue, proposed switching to an async queue with batch writes, and owned the implementation. I ran A/B tests in staging for a week before rollout. We brought latency down to 1.4 seconds and eliminated the timeout errors that had been affecting about 3% of requests."Question 2: Handling technical disagreement
"Describe a time you disagreed with a technical direction your team or manager had chosen. What did you do and how did it resolve?"
What you're evaluating: Whether they can hold a technical position under social pressure, advocate with evidence rather than emotion, and accept outcomes gracefully when overruled.
Weak answer
"I usually go along with what the team decides. I'll share my opinion but I don't like to create conflict."Strong answer
"My tech lead wanted to use a third-party service for our auth layer. I thought we were adding vendor risk for a problem we could solve in a day with a battle-tested open-source library. I wrote a short doc comparing both options on setup time, maintenance overhead, cost, and lock-in risk, and shared it before the next sprint planning. We ended up going with the open-source option. I'd frame it less as winning an argument and more as making sure we'd actually looked at both sides before committing."Question 3: Learning pace
"What's the most significant technical skill you've built in the last 12 months? Walk me through how you built it."
What you're evaluating: Whether learning is intentional and self-directed, or reactive and passive. Strong engineers have a deliberate learning practice and can show how new skills get applied to real work.
Weak answer
"I've been learning Rust. I did some tutorials online and I'm hoping to use it at some point."Strong answer
"I spent the first quarter getting properly comfortable with distributed systems patterns — specifically the tradeoffs between eventual consistency and strong consistency in our Postgres setup. I read the Designing Data-Intensive Applications book, then applied it directly by redesigning how we handle multi-region writes. That work cut our cross-region conflict rate by 80% and I documented the approach so the rest of the team could reason about it the same way."Question 4: Building from scratch
"Tell me about a marketing channel or campaign you built from zero. What did you start with, what did you ship, and what were the results?"
What you're evaluating: Execution at a startup requires building, not managing. This question distinguishes people who have operated a playbook from people who have written one. Look for clear ownership, a before/after, and specific numbers.
Weak answer
"I ran our content programme at my last company. We published 3 to 4 posts a week and grew traffic significantly over the year."Strong answer
"When I joined, there was no SEO strategy at all. I spent the first three weeks auditing the site and doing keyword research to find where we had a realistic shot at ranking. I built a 6-month editorial calendar around 40 target keywords, hired two freelance writers, and set up our publishing workflow in Notion. Within 6 months we went from under 200 organic visits a month to 4,200. That channel now drives about 30% of our inbound pipeline."Question 5: Measurement and accountability
"Walk me through how you measure whether a campaign worked. Give me a real example of a campaign you ran, the metrics you tracked, and what the data told you."
What you're evaluating: Marketing accountability. Many marketers track activity (posts published, emails sent) rather than outcomes (pipeline generated, CAC, conversion rate). Strong candidates connect marketing work to business results.
Weak answer
"We track opens and clicks on our emails, and we look at impressions and engagement on social. I think the campaign did well because our followers grew."Strong answer
"For our last product launch I set up a dedicated landing page with UTM tracking so I could see exactly where signups were coming from. The email sequence had a 34% open rate and drove 180 signups. LinkedIn drove another 60. Paid came in at a $42 CPA compared to our $90 target. The one channel that underperformed was the partner newsletter, which drove 8 signups at about $200 each. I killed that for the next launch and reallocated the budget to LinkedIn."Question 6: Communicating up
"How do you keep a founder or CEO updated on marketing without over-reporting? What does that rhythm look like, and what do you actually share?"
What you're evaluating: Marketing managers at early-stage startups often report directly to founders who don't want to be buried in data. The best candidates share the right signal, proactively flag risks, and don't need to be chased for updates.
Weak answer
"I send a weekly email with all the key metrics. The CEO can look at whatever they need."Strong answer
"I send one paragraph every Friday. It covers the one metric that most reflects whether marketing is working that week, one thing that worked and why, one thing that didn't and what I'm doing about it, and one thing I need from the founder to unblock something. That's it. If there's nothing material to flag, I say so. I save the full dashboard for monthly reviews where we look at trends rather than one-week snapshots."Question 7: Prospecting approach
"Walk me through exactly how you research and prioritise a prospect before you reach out. Use a specific example from a recent outreach."
What you're evaluating: Whether they have a repeatable, personalised prospecting system or spray-and-pray. The specific example tells you whether the process is real or just described.
Weak answer
"I look at their LinkedIn profile and try to personalise the message a bit before sending."Strong answer
"Before I reach out I look for three things: a trigger event, a specific problem signal, and a connection to our ICP criteria. For one prospect last week, I saw they'd just raised a Series A, were hiring for two sales roles on LinkedIn, and their G2 reviews mentioned slow proposal turnaround. I referenced all three in my first line. That email got a reply within 2 hours. I prioritise accounts by intent signal first, then by fit. Cold accounts without any signal go on a lower-priority sequence."Question 8: Handling a prolonged slump
"Tell me about a period where you weren't hitting your numbers. How long did it last, what did you do, and how did it end?"
What you're evaluating: Resilience is the most predictive trait for long-term SDR performance. Strong candidates regulate their own state, diagnose what changed, and adjust without waiting for a manager to intervene.
Weak answer
"Q3 was tough for everyone on the team. The market was slow and our product wasn't quite ready for the segment we were targeting."Strong answer
"I had a 6-week stretch where my connect rate dropped by about 40%. I sat down and pulled my call data to figure out whether it was timing, messaging, or ICP. Turned out I'd drifted toward a segment that had longer sales cycles and the contacts I was reaching weren't the real decision-makers. I rebuilt my sequence around a slightly different title and tightened the ICP criteria. By week 8 I was back above target. I also wrote up what I found and shared it with the team, because two other reps were having the same issue."Question 9: Coachability
"Tell me about feedback you received that you initially disagreed with. What was it, what did you do with it, and what happened?"
What you're evaluating: Whether they can separate ego from performance. Coachable reps improve fast; uncoachable reps plateau and blame their tools. Look for genuine reflection, not performed humility.
Weak answer
"My manager said my emails were too long. I disagreed at first but I shortened them anyway. They seemed to do fine after that."Strong answer
"My manager told me I was pitching too early in discovery calls, before I'd established enough of a problem. I thought I was being efficient. But she had me listen to a recording of one of my calls and I could hear exactly where the prospect disengaged. It was uncomfortable to hear but she was right. I redesigned my discovery framework with more open questions in the first half and shortened my pitch to two sentences. My conversion from discovery to demo went from 22% to 38% over the next quarter."Question 10: Building systems
"Tell me about a process or system you built that other people now depend on. What was broken before, what did you build, and how do you know it's working?"
What you're evaluating: Whether they build for leverage or just execute tasks. Ops hires at startups need to create infrastructure that reduces founder involvement, not require it.
Weak answer
"I set up a project management system in Asana for the team. People use it now and it helps keep things organised."Strong answer
"When I joined, the onboarding process for new hires was entirely in the head of the CEO. Each new hire had a different experience and it was taking about 8 hours of founder time per hire. I spent two weeks shadowing the CEO through an onboarding, documented every step, and rebuilt it as a Notion playbook with checklists, pre-recorded Loom walkthroughs, and a 90-day milestone tracker. The next four hires were onboarded without the CEO being involved in any of the process steps. Time to productivity went from 6 weeks to 3 weeks on average."Question 11: Operating under ambiguity
"Describe a time you had to make a significant decision without enough information. What was the decision, how did you approach it, and what did you learn?"
What you're evaluating: Judgment in grey areas. Ops and CoS roles at startups require acting on incomplete information daily. You want someone who can make a reasoned call, communicate their reasoning, and update based on new information.
Weak answer
"I try to gather as much information as possible before making decisions. I don't like to rush into things without being confident."Strong answer
"We had a supplier relationship that was deteriorating and I needed to decide whether to put a backup vendor in place before our busy season without knowing whether the main supplier would actually fail. I had three weeks to act. I wrote a quick decision brief that laid out the cost of being wrong in each direction: switching early at $40k in setup cost vs. being stuck without supply at peak season with an estimated $300k revenue impact. I recommended we set up the backup. The main supplier ended up being fine, but the decision was still correct given what we knew at the time. I wrote that into the brief upfront so the rationale was transparent."Every structured question above becomes twice as powerful with one follow-up: "What would you do differently if you were in that situation again?"
This question surfaces self-awareness. A candidate who gave a strong example but can't identify any improvement has limited growth potential. A candidate whose initial answer was average but who articulates exactly what they'd change differently shows the kind of reflection that compounds over time.
The follow-up hierarchy
Use follow-up questions consistently across all candidates. If you probe one candidate deeper than another, you've introduced inconsistency into your process and the comparisons break down.
Questions without a scoring framework are half a structured interview. Once you have the questions set, define what each score level looks like before you run any interviews.
Scoring Guide: What Each Level Looks Like
| Score | What the answer contains | Common signal |
|---|---|---|
| 4: Exceptional | Specific example, clear ownership, measurable outcome, reflection on what they'd improve | You lean forward. You want to hear more. The answer changes your view of the candidate. |
| 3: Strong | Specific example with clear ownership and a concrete outcome. Minor gaps in depth or reflection. | Solid evidence of the competency. This candidate meets the bar. |
| 2: Weak | Vague example, shared ownership, no measurable outcome, or a hypothetical answer to a behavioral question | You're not sure the competency is there. You'd want a follow-up to confirm. |
| 1: Absent | No relevant example. Generic statements. Attribution of results to luck, the team, or external factors. | Clear gap. This isn't about inexperience, it's about pattern. A score of 1 on a must-have competency is disqualifying. |
Score independently immediately after the interview, before any group discussion. Once scores are submitted, share them simultaneously in the debrief rather than sequentially. Whoever scores first anchors everyone else.
For a full structured scoring template tied to these questions, see the free interview scorecard template or the interview scorecard examples by role.
HireLikeaPro generates a custom set of structured interview questions tied to the competencies in your job description, plus a scorecard to evaluate every answer. Free forever, no credit card.
Build My Interview Questions Free →