11 Strategic Planning Workshop Exercises That Work

Most lists of planning exercises are lists of icebreakers. These are not. Every exercise below exists to force a specific decision that groups otherwise avoid, and each one names the failure it prevents.

Pick four or five. A day that runs eleven exercises has run none of them properly.

Before you pick anything: what are you stuck on?

Choose by the problem, not by the exercise:

The problem	Use
Everyone agrees, suspiciously	Pre-mortem · Disagree and commit
Too many priorities	Forced ranking · The capacity check
No shared picture of reality	Timeline of the year · Stop / start / keep
One loud voice dominating	Silent writing · Dot voting
Vague ownership	Owner naming · The Tuesday test
Vague measures	Measure sharpening · The 90-day proof

1. Silent writing (5–10 min)

Prevents: anchoring on whoever speaks first.

Ask the question. Everyone writes their own answer alone, on paper, before anyone speaks. Then each person reads theirs aloud with no discussion until all have read.

This is the highest-leverage facilitation move available and it costs five minutes. In any group with a status gradient — which is every leadership team — the first spoken answer sets the range for everything after it. Silent writing removes that.

Use it before: the North Star discussion, priority generation, any question where the CEO has a known view.

2. Stop / start / keep (30 min)

Prevents: planning as pure addition.

Three columns. Groups of three, 15 minutes, then report back.

The trick is a hard rule: the stop column must have at least as many items as the start column. Without it, teams fill start generously and stop with one item nobody cared about, then wonder in March why nothing has capacity.

3. Timeline of the year (30 min)

Prevents: arguing about last year from different memories.

Draw the last twelve months across a wall. Everyone adds events where they happened — launches, departures, wins, the month everything broke.

The value is in the disagreements. When two people place the same event three months apart, or one person's disaster is absent from everyone else's timeline, you have found a gap in shared understanding that would otherwise distort the whole planning conversation.

4. The one number (60 min)

Prevents: a plan with no arithmetic behind it.

Silent writing → read aloud → cluster on a wall → converge.

Budget a full hour and expect real conflict. This disagreement exists in your company already; the only question is whether it gets resolved here or quietly, in a hundred separate decisions, over the next year.

If you genuinely cannot converge, that is a finding, not a failure — and it means the session's job changed. Stop planning and resolve the direction question first.

5. Forced ranking (45 min)

Prevents: the twelve-priority "plan."

Two axes only: impact on the number, effort to deliver. Place every candidate. Then draw a line and cut everything below it, out loud.

Keep the cut list. Written down, visible, carried into the plan document. A cut list that vanishes gets quietly re-adopted by individuals over the following quarter, which is how a team ends up delivering twelve priorities it never agreed to.

Resist adding a third axis. Every axis you add moves the decision further from judgement and closer to arithmetic that feels objective and isn't.

6. Dot voting (15 min)

Prevents: convergence theatre, where the group agrees because disagreeing is awkward.

Each person gets three dots and places them on the wall. Vote simultaneously — count down from three and everyone places at once, so nobody is following anyone.

Fast, visual, and it makes the shape of disagreement legible. A candidate with two dots from two people is not the same as one with four dots from one person, and the wall shows you which you have.

7. The capacity check (20 min)

Prevents: the plan that was never physically possible.

For each surviving priority, ask the owner one question: what would you have to stop doing to deliver this?

If the answer is "nothing, we'll absorb it," you have found a priority that will not happen. Absorbed work is work that gets dropped the first week something urgent lands.

8. Owner naming (20 min)

Prevents: ownership by department.

Go priority by priority. Name a person, aloud, and have them accept it aloud. Two rules: one human per priority, and the owner must be present in the room.

Naming a department is the single most common way a plan quietly fails. "Marketing owns this" survives the workshop and dies on contact with any week that has other work in it.

9. Measure sharpening (30 min)

Prevents: measures nobody can compute.

For each priority's proposed measure, ask three questions:

On 31 December, will this be unambiguously true or false?
Who computes it, and from what source?
Could someone hit this number and still have failed at the intent?

Question three catches the gameable measures. "Ship the new onboarding" is satisfied by shipping something bad. "40% of new accounts reach activation within 7 days" is not.

10. Pre-mortem (45 min)

Prevents: politeness suppressing real objections.

"It is twelve months from now. The plan failed. Why?"

Silent writing first, then round the room. This is the highest-yield exercise in the whole day, for a reason that is entirely social: it is far easier to describe a hypothetical failure than to object to a colleague's plan in front of the CEO. The pre-mortem gives people permission to say the thing they have been sitting on since 10am.

Capture every cause. The two or three most-mentioned become named risks with owners.

11. The Tuesday test (15 min, closing)

Prevents: a plan that never reaches anyone's actual work.

Pick a real person who was not in the room. Ask: what does this plan change about their Tuesday?

If nobody can answer concretely, the plan stops at the leadership layer. That is not a communication problem to solve later — it is a cascade problem to solve now, while everyone is still in the room.

Two exercises to avoid

SWOT. It generates four lists and no decisions. Teams enjoy it because it feels productive and commits to nothing. If you want the same inputs, use stop/start/keep, which at least ends in a trade.

Anonymous digital polling for the big call. Anonymity is right for surfacing dissent and wrong for commitment. People need to hear each other say the number out loud, because a decision nobody voiced is a decision nobody owns.

Running the whole session

These exercises assume a structure around them. The full one-day planning workshop agenda sets out the timings and the preparation the day depends on — and for a shorter session, the annual planning meeting agenda compresses this into two to three hours.

If the honest problem is that you don't have the inputs yet — no agreed number, no KPI structure below the leadership team — no exercise fixes that in a morning. That is the gap our free planning workshop is built to close.