"The best screening questions are specific to the role, tied to something the candidate has actually claimed, and designed to be hard to rehearse." Four types of questions consistently reveal signal: scenario questions grounded in the role's real challenges, probing follow-ups to the candidate's own claims, questions tied directly to stated accomplishments, and questions that test for adjacent skills the role requires. Generic questions — *tell me about yourself, why are you looking to move, what's your biggest weakness* — produce rehearsed answers and reveal almost nothing.

AI makes it possible to generate a tailored question set for any role in under five minutes. What follows is the framework, the four types, and the prompt pattern to produce them on demand.

Why Most Screening Questions Fail?

Walk into any recruiting team's shared drive and you'll find a document called something like "Phone Screen Questions" that hasn't been updated in two years. It contains the same twelve questions that get asked of every candidate for every role: "Walk me through your background. Why are you looking for a new opportunity? What are you looking for in your next role? What's your biggest strength? Biggest weakness?"

Every candidate who has interviewed more than twice in their adult life has heard these questions. Every candidate has a rehearsed answer. The rehearsed answer tells you what the candidate wants you to know, not what you need to know about them. You finish the call feeling like you learned something. You didn't. You collected twelve polished monologues.

This is the core failure mode of screening. A question that's easy to prepare for is a question that produces preparation, not truth. The goal isn't to trick candidates — the goal is to ask questions that can't be rehearsed because the answers depend on the specific role, the specific situation, or the specific claim the candidate has made.

What Makes a Screening Question Useful?

A useful screening question has three properties.

It's specific to the role. A screening question for a Senior PM at a Series B fintech should be different from a screening question for a Senior PM at a Series D consumer company. The challenges are different. The relevant experience is different. A generic PM question fits neither context and reveals nothing about fit for either.

It can't be fully prepared for. The best questions require the candidate to respond to something they haven't seen before — a specific scenario, a probing follow-up to their own answer, a detailed question about a claim on their resume. Preparation helps. It doesn't substitute for the underlying experience.

It produces a signal that discriminates. A useful question produces different answers from strong and weak candidates. A weak question — "Tell me about a time you failed" — produces similar-sounding answers from almost everyone, because the answer pattern is well-known. A strong question produces answers that vary meaningfully based on what the candidate has actually done.

If a question doesn't hit all three properties, it's theatre.

The Four Types of Screening Questions That Work

Type 1: Scenario questions grounded in the role

A scenario question describes a specific, realistic situation the person would face in the role, and asks how they'd approach it. "Your company just acquired a smaller competitor. You're the PM for the product category that now overlaps. Walk me through how you'd handle the first 60 days."

Notice what this does. It's tied to a real situation senior PMs face. It requires the candidate to reason about priorities, stakeholders, and trade-offs — not recite a framework. It's hard to prepare for because the scenario is specific to your role at your company stage. Candidates who've operated at this level will have thoughts. Candidates who haven't will pattern-match to generic advice.

These questions require thought to design. Each one should reflect a situation the candidate would actually encounter. Generic scenarios ("how would you handle a disagreement with a stakeholder") don't discriminate; specific ones do.

Type 2: Probing follow-ups to the candidate's own answers

This is the highest-value technique in screening, and it's the one most recruiters skip.

A candidate says: "I led the migration of our data platform to Snowflake, saving the company 40% on infrastructure costs."

A weak recruiter moves to the next question. A strong one asks: "What was the hardest decision you had to make during that migration, and what did you decide?"

The follow-up reveals whether the candidate actually led the migration or merely participated in it. Someone who led it will have a textured answer — a specific decision, specific trade-offs, specific regret. Someone who participated will have generalities or deflect to the team.

Every claim on a candidate's resume deserves at least one probing follow-up during screening. Not as an interrogation — as a way to understand what the claim actually represents. AI can help you generate these follow-ups in advance, tailored to the specific resume you're screening.

Type 3: Questions tied directly to stated accomplishments

Closely related to probing follow-ups, but distinct enough to name separately.

These are questions that take a specific claim from the resume and ask the candidate to walk through it in detail. "You said you scaled the sales team from 5 to 30 at [Company]. Walk me through the first three hires you made and why you chose them."

This question is unreachable by preparation in any useful way, because the answer requires the candidate to remember specific people, specific reasoning, and specific outcomes. A candidate who did the work will have the answer in reach. A candidate who inflated the claim will produce fog.

The rule of thumb: for any quantified accomplishment on a senior-level resume, generate one question that requires walking through the details. Two or three such questions in a 30-minute screen will tell you more than ten generic behavioral questions ever could.

Type 4: Questions testing for adjacent skills the role actually needs

Most roles require a set of adjacent skills that aren't in the job title but are essential to success. A Senior PM role at a startup requires not just product management but also some amount of hands-on design instinct, enough sales literacy to coach the go-to-market team, and executive stakeholder management. The JD rarely lists these explicitly. The role fails without them.

Screening questions should probe for these adjacent skills. "Walk me through the last time you had to get a reluctant sales leader to change their approach. What was the argument, and how did you run it?" That's a sales-literacy question dressed up as a situation. The answer tells you whether the PM can actually operate with a go-to-market partner — which is more than half of what the role is.

Identify the three or four adjacent skills your role will actually require, and design at least one question per skill. Most screens never touch these areas and instead ask the candidate to describe their product management philosophy for the fourth time.

The Prompt That Generates a Tailored Question Set

You don't need to build this alone. Once you've done the thinking about what the role actually requires, AI generates a complete screening question set in a few minutes.

"I'm screening candidates for a [ROLE] at a [COMPANY STAGE + INDUSTRY]. The core challenges of this role are: [LIST 3–4 CHALLENGES]. The adjacent skills required beyond the core function are: [LIST 2–3 ADJACENT SKILLS].

Generate 8 screening questions for a 30-minute phone screen:
3 scenario questions grounded in the specific challenges above
2 adjacent-skill questions dressed as situations
3 "probing template" questions I can adapt on the fly to whatever accomplishments appear on the candidate's resume

For each question, include: what it's testing, the signal I should listen for in a strong answer, and a red flag in a weak answer. "

Run this for your next open role. Compare the question set it produces to the generic list your team has been using for two years. The difference is what better screening looks like.

Why This Matters More Than it Used To?

Screening used to be filter work. Get rid of the obvious mismatches, send the rest to the hiring manager. That was fine when application volumes were lower and candidates hadn't figured out how to optimize their resumes to pass keyword filters.

Neither of those conditions holds now. Inflated resumes clear the first filter routinely. The phone screen is often the first point at which signal and claim diverge — and the only way to catch it is questions designed for that job. Generic questions collect generic answers and pass inflated candidates through to hiring managers, who discover the problem two rounds later.

Better questions aren't a nice-to-have. They're the specific place in the process where quality of hire gets won or lost.

The [Prompt Pack](./lesson-21-prompt-pack) includes ready-to-use screening question prompts for fifteen common role types.

Next: Lesson 14 — Running Interviews That Actually Reveal Something