Methods

How to run a surface analysis workshop in 90 minutes

By Idris Achterberg

What you need

A room, a wall you can write on, and a team willing to admit their plan rests on something. Ninety minutes. No prior geometry required.

Surface analysis is assumption mapping done as a group exercise. The goal is to make visible which beliefs are load-bearing, which are decorative, and which are silently bending under weight no one has measured. Most strategy workshops produce aspirational slide decks. This one produces a map you can actually use on Monday.

The setup (10 minutes)

Pick one initiative. Not the whole strategy — one discrete plan with a clear intended outcome. A product launch. A pricing change. A new content calendar. Something the room has collectively decided to do.

Write the initiative at the center of the wall. Draw a horizontal line beneath it. That line is the surface: the set of assumptions this plan is resting on. Everything below the line is foundational. Everything above is what you're trying to build.

Tell the room: we are going to map what has to be true for this to work.

Round one: list the assumptions (20 minutes)

Go around the room. Each person names one assumption the plan depends on, and you write it below the line. No debate yet. No "actually that's not an assumption, that's a fact." If someone believes the plan requires it, it goes on the wall.

Common examples:

  • "Our users check email daily."
  • "The sales team will adopt the new CRM."
  • "Competitors won't react for at least two quarters."
  • "Leadership will still care about this in six months."

Keep going until the room slows down. You want 12–20 assumptions. Fewer means the team is being polite. More means you need to cluster.

Round two: weight and curvature (30 minutes)

Now the real work. For each assumption, ask two questions:

  1. How much weight is this carrying? If this assumption is wrong, does the plan limp forward or collapse entirely? Mark high-load assumptions with a double underline.
  2. How confident are we? Not "do we hope this is true" — do we have evidence, or are we guessing? Assumptions you're guessing about get a question mark.

The dangerous ones are high-load and low-confidence. A plan resting on those is curved in a way that will surprise you later. One client — a mid-size software team shipping a new dashboard — discovered they'd built everything on the assumption that their enterprise users had reliable internet at remote sites. No one had checked. The surface was not flat. The plan did not ship on time.

Round three: test three assumptions (20 minutes)

You will not have time to validate everything on the wall. Pick three:

  • The highest-load assumption
  • The least-confident assumption
  • The assumption someone in the room is visibly uncomfortable about

For each one, decide: what is the fastest, cheapest way to check if this is true? Not "run a six-month study." What can you do this week?

Examples:

  • Call five customers and ask directly.
  • Pull login data from the last 90 days.
  • Spend one day with the team that has to execute this.

Write those tests on the wall next to the assumptions. Assign owners. Set a deadline that is not "sometime next quarter."

Round four: the question no one wants to ask (10 minutes)

Now ask the room: if we ran these three tests and all three assumptions turned out to be false, would we still do this?

Sometimes the answer is yes. The plan has value even if the surface is shakier than hoped. Sometimes the answer is no, and you have just saved the team six months of work they would have regretted. Sometimes the answer is "we'd do a different version," and that different version is now visible.

This is the moment when someone will say, "But we've already committed to this." That is not a reason to skip the test. That is a reason to run it faster.

What happens after

Someone — usually you — takes a photo of the wall and turns it into a document. Not slides. A literal list: these are the assumptions, these are the tests, these are the owners, these is the deadline. Circulate it within 24 hours.

Two weeks later, check in. Did anyone run their test? If yes: what did you learn, and does the plan need to change? If no: why not, and is this plan actually a priority?

A surface analysis workshop does not guarantee your strategy will work. It guarantees you will know which assumptions you are betting on, and it will be your fault if you do not check them. Most teams never get that far.

A note on the geometry

Some surfaces are genuinely flat. Most are not. If your plan depends on customer behavior, market conditions, or internal adoption, the surface has curvature you cannot see from directly above. This is not a metaphor. A curved surface does not support weight the same way in all directions. The plan that looks stable in a slide deck may tip under load in the field.

The workshop will not tell you the exact shape. It will tell you where to measure.

← All insights