Skills Assessment: The Questions That Actually Work

There is a difference between someone who knows the answer and someone who can do the thing. Most assessment questions only test the first one.

This guide is about designing questions that get you to the second - practical, role-relevant, and hard to bluff.

---

What Are You Actually Trying To Find Out?

Before you write a single question, be clear about what you need to know. A useful assessment should tell you three things:

• What someone already knows and can demonstrate
• What they think they know but can't apply
• What they don't know they're missing

Most assessments only ever surface the first. The questions below are designed to get to all three.

---

Question Types And What They Reveal

The type of question determines the depth of answer. Most assessments stay at the top of this table.

Question Type	Example	What It Reveals
Recall	What does GDPR stand for?	Whether someone has been exposed to the topic
Understanding	Why does GDPR matter for how we handle customer data?	Whether they've made sense of it
Application	How would you handle a data subject access request?	Whether they can use it in context
Analysis	What could go wrong in our current process?	Whether they can think critically about it
Demonstration	Walk me through what you'd do if a customer requested their data	Whether they can actually do it

A few things worth knowing:

• Recall questions are the easiest to ask and the least useful on their own. Knowing what something is called does not mean someone can do anything with it.
• Application and analysis questions reveal the gap between people who have genuinely understood something and people who have just sat through it.
• Demonstration questions are the hardest to scale but produce the most reliable evidence of capability.

---

Before You Finalise Any Question

• Does it require the person to do something with their knowledge, not just recall it?
• Is it specific enough that a vague answer would be obviously insufficient?
• Could someone answer it correctly without actually understanding the topic?
• Does it relate to something they would encounter in their actual role?
• Would the answer tell you something you could act on?

If a question fails more than one of those, it probably is not earning its place.

---

Template: Question Sets By Level

Use these as a starting point and adapt to your topic. The goal is to move through levels - starting with recall to establish a baseline, then pushing towards application and demonstration to find out what someone can actually do.

Recall

• What is [topic]?
• Can you name three examples of [concept]?
• What are the key rules or requirements around [area]?

Understanding

• Why does [topic] matter for our team specifically?
• What's the difference between [concept A] and [concept B]?
• What would happen if we didn't follow [process]?

Application

• How would you handle [scenario]?
• What would you do first if [situation] happened?
• Talk me through how you'd approach [task].

Analysis

• What could go wrong with our current approach to [topic]?
• Where do you think the biggest risk is in how we currently handle [area]?
• What would you change and why?

Demonstration

• Show me how you'd [task].
• Walk me through your process for [scenario].
• If a colleague asked you to explain [concept], what would you say?

---

The Follow-Up Questions That Do The Real Work

The most revealing moment in any assessment is often the follow-up - the question you ask when someone gives an answer that sounds right but feels thin.

A few that work across most topics:

• Can you give me an example from your own experience?
• What would you do if that approach didn't work?
• How confident are you in that answer?
• What would you want to check before acting on that?

These are harder to build into a survey. They are one of the strongest arguments for having a real conversation.

See also: Do You Actually Know What Your Team Can Do?