All methods
StrategyIntermediate

Goals, Signals, Measures

Goals, Signals, Measures is a 90-minute workshop from the Atlassian Team Playbook that defines what success looks like before a project starts. The team agrees up to three outcome goals, names the signals that would show movement toward each one, and turns those signals into specific, time-bound measures. It produces a short success definition that can be checked later instead of argued about.

Duration
2h–2h
Group size
6–8 people
Materials
Whiteboard or digital board, Sticky notes, Markers…

Facilitation script

  1. 1

    Set the stage. The project owner connects the project to the wider mission in a few minutes; the group asks questions until the purpose is clear.

    10 min
  2. 2

    Silent goal brainstorm on sticky notes, one goal per note, written as an outcome. Post them as they are written.

    15 min
  3. 3

    Sort into No, Maybe and Yes as a group, cut the Yes column to three goals at most and name an owner for each.

    10 min
  4. 4

    Signals. Take one goal at a time and collect three to five signals that would indicate progress. Keep them observable.

    15 min
  5. 5

    Measures. Convert signals into specific, ambitious, time-bound measures with a number and a date. Rewrite any yes/no measure.

    30 min
  6. 6

    Wrap up. Read each chain of goal, signals and measures aloud, sanity-check the timelines, and assign the data-gathering tasks.

    10 min

Tips

  • Most groups write outputs first, such as 'launch the new onboarding'.

  • Ask 'and what changes for whom if that ships?' until the note describes an outcome.

  • Hold the limit of three goals firmly, because the measures step takes a third of the session and falls apart with five.

  • Before closing, ask of every measure where the number will come from and whether a baseline exists today.

Common pitfalls

  • Accepting deliverables as goals, which leaves the team measuring whether it shipped rather than whether anything changed

  • Keeping five or six goals to avoid a hard conversation, so the measures step runs out of time and the last goals get vague targets

  • Choosing measures because the data is easy to get, which produces numbers that move without telling you anything about the goal

  • Writing yes/no measures such as 'feature adopted', which cannot show partial progress and hide a near miss as a failure

  • Leaving without owners for data collection, so the first check-in weeks later finds that nothing was instrumented and no baseline exists

Variations

For a remote team, run it on a digital whiteboard with one column per goal and three rows for goal, signals and measures. If the group is larger than eight, split into sub-groups by goal for the signals and measures steps and review together at the end. Teams that already use OKRs can run the first four steps as preparation and write the measures directly as key results.

Where it fits

Project kick-offShifting a team from outputs to outcomesChoosing success metrics for a new feature or initiativeAligning a project team with its sponsor on what done meansPreparing the input for a later OKR session

When to use it

  • A project is about to start and the team can list its deliverables but not what should be different afterwards

  • Sponsor and team would give different answers to the question of how success will be judged

  • Past projects were declared successful because they shipped, with no evidence of effect

  • The team wants a few metrics for one project without adopting a full OKR system

  • A dashboard exists with dozens of numbers and nobody can say which ones matter for this project

When not to use it

  • You are setting quarterly objectives across a department or company; an OKR Workshop handles alignment between teams, which this session does not

  • The product has no agreed overall direction yet; a North Star Metric Workshop comes first, and project goals follow from it

  • The work is a fixed-scope delivery with contractual acceptance criteria and no room to act on what the measures show

  • The project owner is absent or cannot explain why the project exists; the first step depends on that link to the wider mission

Related methods

Frequently asked questions

What is Goals, Signals, Measures?▾

It is a workshop from the Atlassian Team Playbook for defining project success up front. The team agrees up to three outcome goals, lists the signals that would show progress toward each, and converts those signals into specific, time-bound measures.

What is the difference between a signal and a measure?▾

A signal is an observable sign that you are moving toward the goal, described in words, such as new users finishing setup without contacting support. A measure makes that sign countable and gives it a target and a date, such as the share of new users completing setup unaided reaching a stated level by the end of the quarter.

How long does it take and who should attend?▾

The session runs about 90 minutes, with around 15 minutes of preparation. Atlassian suggests six to eight people: the project owner, the core team, and anyone whose data or sign-off the measures depend on.

How does it differ from OKRs?▾

OKRs are a recurring goal-setting system across an organisation, with objectives and key results reviewed on a cadence. Goals, Signals, Measures is a single workshop for one project, and its extra middle step, signals, helps a team reason from a goal to a metric. The output can be translated into OKRs if the organisation uses them.

Why limit the session to three goals?▾

Each goal needs several signals and at least one well-formed measure, and that work takes most of the session. With more than three goals the measures become rushed and the team ends up tracking more numbers than it will look at.

🪡

Plan your next workshop with AI

Workshop Weaver helps you combine methods like Goals, Signals, Measures into a complete, timed agenda in minutes.

Try it free

Method descriptions on Workshop Weaver are original content written by our team, based on established facilitation practices. This method was inspired by work from Atlassian Team Playbook.