GOOSE stands for Goals, Outcomes, Opportunities, Solutions, Experiments. It’s a structured framework to help individuals and teams navigate uncertainty, make better decisions, and continuously learn to have a real impact. By applying systematic learning at every stage, GOOSE ensures that effort is aligned with meaningful goals, the right problems are addressed, and solutions are validated with evidence.
We are heavily inspired by
Design Thinking https://www.interaction-design.org/literature/topics/design-thinking and Teresa Torres’ Opportunity Solution Trees https://www.producttalk.org/opportunity-solution-trees/.
Systematic Learning
A common thing in every step of the way is making sure that we know what we need to learn and then actually learning it. This is done by organizing our knowledge.
How do we organize our knowledge?
At every step of the way, we should recognize what we think we know and what we need to find out. As we are about to research or researching a topic, we should organize our knowledge:
Known knowns (assumptions): Things we believe to be true. These need to be recognized and validated as necessary.
Known unknowns (questions): Gaps in our knowledge that require research to answer.
Unknown unknowns (discoveries): Insights we can only uncover by staying curious, open-minded, and alert.
Ultimately we are making educated guesses, but the key to great educated guesses is educating ourselves.
The diverge/converge diamond
Also, at every step of the way, we should be aware of the options we have and choose the best one. This means we should focus on both generating options and figuring out the best one:
Generate options (diverge): This is where we need to explore alternatives and have an opportunity to innovate.
Choose the best option (converge): We can’t do it all at once, so for now, just choose one or the best few.
Why is this important?
This rhythm of diverging and converging is the heart of modern problem-solving. It is used in every phase of the GOOSE model:
Goal: Ideate many possible goals → choose the most important one.
Opportunity: Discover many customer problems → choose the one with the biggest impact.
Solution: Ideate many solutions → choose the most promising one to experiment.
If we only diverge, we create chaos. If we only converge, we create tunnel vision and make bad decisions. When we do both, we get results.
Goal
Outcome
Opportunity
Opportunity
Solution
Solution
Experiment
Experiment
1. Goals – Choosing the Right Direction
The first step is identifying the right goal to pursue. Goals provide clarity of direction and purpose, similarly to an Objective in the OKR framework.
Goals should define the purpose of the work and be inspirational, but also be authentic. They will guide all the work in the later steps, meaning inauthenticity will cause misunderstandings and wasted effort.
Goals might be business centric for example “to be the biggest x” or product centric for example “to be the most flexible”. Ultimately product goals should link to business goals and teams should not try to directly pursue business goals.
Business goals might be the “ultimate goal” but product goals are the strategy we have chosen to pursue them.
How to choose a goal:
List many possible goals (diverge)
We can start with something simple (to be the best x in y), but we’ll keep adding options until we are satisfied
Organize your (lack of) knowledge about the goals and research to gain a better understanding.
Does this make sense for the business? Would it provide value for the customer? Is it achievable?
Choose the best goal (converge)
Stick with the goal for a while, but don’t get married to it. Keep refining it as you go and learn more
Keep learning more about it
Goal clarity workshop
To transform the company's goals into testable strategic hypotheses.
2. Outcomes – Measuring Progress
Once a goal is defined, the next step is determining what measurable outcomes would demonstrate progress. These map to the Key Results in OKRs.
Researching the goal should yield insights about what outcomes are likely to drive progress the most.
Outcomes can be business centric for example to “increase the customer lifetime value” or they might be product centric for example to “increase feature x usage“. Ultimately product outcomes should link to business outcomes and teams should not try to directly pursue business outcomes.
New features might improve business metrics, but not directly. A better onboarding flow might directly decrease the time used onboarding or increase the user activation rate (product outcomes) and in addition, indirectly decrease customer acquisition cost (business outcome).
How to choose an outcome:
List many possible outcomes (diverge)
Organize your (lack of) knowledge about the outcomes and research to gain a better understanding.
Where are we now? Where should we be? Can we get there? How much effort would it take?
Choose the best outcome (converge)
Keep learning more about it
Setting product metrics
This guide is designed to help product teams define clear, actionable metrics that can be tested, measured, and aligned with business outcomes.
Defining product outcomes
Learn to identify critical bottlenecks, map dependencies, prioritize high-impact problems, and define SMART outcomes that drive real value and strategic alignment.
3. Opportunities – Finding the Right Problem to Solve
Outcomes describe the impact we want to see. The next step is uncovering opportunities, or the right problems to solve.
Opportunities should be described as something a user might say - for example a streaming service user might say “It’s hard to choose what to watch”.
They should not be “features in disguise” as in, they only allow a single solution. “I want to watch <a show that is not on the service> in the service”.
So an actual pain point the user might have that could be solved in a number of ways.
How to choose an opportunity:
List many possible opportunities (diverge)
Organize your (lack of) knowledge about the opportunities and research to gain a better understanding.
Is this a widespread problem or just a minor thing? Combine qualitative and quantitative research as in other steps.
Choose the best opportunity (converge)
Keep learning more about it
4. Solutions – Choosing the Right Response
With the opportunity framed, it’s time to explore solutions. These can take the form of features, projects, or initiatives.
Sometimes our research directly yields solutions. For example a customer in an interview might request a feature. In these cases it is important to understand the opportunity behind the solution. This way we have to have more options on how we solve it and can potentially generate better outcomes for less.
How to choose a solution:
List many possible solutions (diverge)
Organize your (lack of) knowledge about the solutions and research to gain a better understanding.
Common risks include viability (does this make sense for the business), feasibility (can we make this a reality) and usability (can people use this without an instruction book). These should be verified before committing to delivering a solution.
Choose the best solution (converge)
Keep learning more about it
5. Experiments – Learning What Works
Finally, solutions should not be automatically assumed to be correct. With experiments, we measure the objective effectiveness of a solution and generate lessons learned. Ideally an experiment is linked to the outcomes we are trying to achieve.
Sometimes a solution is assumed not to directly affect the outcome, but still be helpful. Even in these cases we might be able to evaluate the effectiveness of a solution using proxy metrics we assume are linked to the actual outcomes.
Even something as simple as a “how many users actually use this feature” metric might yield unexpected discoveries that help the team learn more.
Define measurable hypotheses
Run experiments (small, fast, and focused where possible)
The more risk we have the smaller the experiment should be
Evaluate results with evidence
Feed lessons learned back into the GOOSE cycle
Do we keep the feature? Maybe change it? Maybe kill it?
This ensures continuous learning and adaptation rather than static execution.
Conclusion
The GOOSE framework provides a repeatable approach to learning and problem-solving:
Goals: Set the right direction.
Outcomes: Define measurable progress.
Opportunities: Find the right problems.
Solutions: Explore and select the best responses.
Experiments: Validate with evidence and learn.
At every stage, GOOSE emphasizes systematic learning and the diverge → converge cycle. This not only helps us to maximize impact but also builds resilience by revealing the unknown.
