Interview Questions Are Mostly Pointless
I've conducted probably two hundred interviews across design, engineering, and product teams over the last decade. The ones that actually revealed anything useful were almost never the polished questions from any of those viral listicles you see on HR blogs. They were the awkward, slightly strange things that made the person stop performing and start thinking out loud. Most interviewers ask questions they already know the answer to and are just looking for a specific response. That's not an interview. That's a fill-in-the-blank worksheet with a salary attached. Creative Interview Questions To Ask isn't about being quirky for its own sake. It's about designing questions that short-circuit rehearsed answers and expose how someone actually works when they don't know what you want them to say.
How to actually design questions that work
Start with the job, not the question. If you're hiring a senior designer, don't ask them to describe their favorite app. Ask them to walk through how they'd redesign the check-out flow for a grocery delivery service where the payment button is broken on mobile. Now you get to see if they think about users, constraints, trade-offs, and whether they push back or just agree. Here's the part nobody tells you: the best questions are usually incomplete. Give the candidate a scenario with missing information and watch what they do. Do they ask clarifying questions? Do they make assumptions and state them? Do they get flustered and default to generic answers? The gap in the brief is the whole point. I learned this the hard way when I once asked a candidate to "design something for grandparents" and they spent forty-five minutes asking about arthritis, vision loss, tech literacy, and whether these grandparents lived alone or with family. They got the job. The three candidates who just started describing UI mockups didn't.
Questions that actually reveal competence
The constraint question: "You have three days to ship a feature that normally takes six weeks. What do you cut and what do you fight to keep?" This one exposes whether someone understands scope, priorities, and their own willingness to compromise. Good answers mention talking to stakeholders. Terrible answers talk about working harder or just grinding it out. The failure breakdown: Not "tell me about a time you failed." That's too broad and invites a humblebrag. Try "walk me through a project that went wrong and tell me the exact moment you realized it." The specificity forces honesty. Most people can't fake that level of without sounding rehearsed. The peer evaluation: "Your teammate keeps making decisions without consulting you. How do you handle it?" This reveals conflict resolution style, ego, and whether they'll escalate properly or go behind someone's back. I've seen this split candidates wide open. The ones who described a direct conversation handled leadership roles better than anyone I'd ever put on a team.
Get the Full Details
The absurd scenario: "A client asks you to make their logo bigger and more colorful because their competitor just did. You think that will make it worse. What do you say?" This tests persuasion, not design skill. Can they articulate why something is bad without sounding arrogant? Can they offer alternatives? Can they read the room? Junior designers often freeze here. Seniors who've survived client work usually have a playbook. The retrospective angle: "What would your last manager say is the one thing you should stop doing?" If the candidate has no answer or blames their boss entirely, that's data. If they name a real pattern they recognize in themselves, you've found someone who reflects instead of deflects.
What to watch for instead of what they say
The content matters less than the process. I sit through these and take zero notes on whether the answer is right. I watch the pacing. Do they pause and think or immediately launch into a prepared story? Do they correct themselves mid-answer? That self-correction is usually more valuable than any polished response. People who revise their thinking in real time are people who actually think. Also note what they don't ask. In a thirty-minute interview, if the candidate hasn't asked a single question by the end, something is off. Either they aren't genuinely interested or they're so practiced at performing compliance that they've forgotten to be curious. Curiosity is non-negotiable for anyone I hire past entry level.
Where this approach falls apart
These questions don't scale well. A single 45-minute deep interview with a creative scenario takes as long as screening twelve candidates with traditional questions. If you're hiring for a company with two hundred open roles, you're not going to use this method for every position. It works best for roles where a bad hire is genuinely expensive — senior contributors, people who'll influence teams, roles where cultural fit is actually about how someone solves problems rather than whether they're likable at lunch. Another limitation: these questions favor people who are comfortable with ambiguity. Some excellent engineers, some excellent accountants, some excellent operators are people who like clear problems with clear answers. They'll look uncertain in these interviews even though they'd be phenomenal at their actual job. I've passed over candidates who were clearly brilliant but just couldn't perform well under open-ended scenarios. Don't let this method become its own bias. If you're in a volume hiring situation, a modified approach works better. Instead of full creative scenarios, use one or two focused constraint questions per candidate and pair them with a practical skills test that's closer to actual work. The skills test tells you whether they can do the job. The one or two weird questions tell you whether they'll figure things out when the job description turns out to be wrong.
The resource I keep coming back to when building out question banks is the actual job itself — the daily problems your team faces. The next time someone complains about a recurring issue in standup, that complaint is material for an interview question. Real problems make real questions. Everything else is just theater.
