How product teams can measure progress through outcome OKRs

Most product teams review goals in hindsight to see whether they have succeeded. But what if you could not just assess what you’ve already done but prioritize your actions along the way? To help do that with a strong focus on outcomes, Product Management coach and author Tim Herbig took ProductTank Amsterdam through a hands-on workshop. He provided context behind the idea of OKRs and surprised some of the other attendees with new thoughts on the matter.

The session was a mixture of both theory and practical exercises for the audience to directly try out the introduced concepts and get feedback from Tim and the community. Tim closed the workshop by discussing questions from the audience. Below is a summary of the workshop’s takeaways:

“Outcome over output” is a well-known claim in product management. Output OKRs (e.g. “feature x has been released”) are said to be rather shallow, not inspirational and are likely to block learning opportunities. However, there is nothing wrong with output OKRs per se for measuring success. The team should just be aware of the consequences when they settle on a specific type of key result. The key results should give a holistic picture of the objective, capturing the objective from as many perspectives and levels as possible. This will give more space to come up with the best solution(s), but also requires more confidence and skill from the team.

Based on the consequences of the key results, OKRs can be separated into three different types:

Categorising your OKRs like this will likely not be enough for actually working with them on a day-to-day basis. A second layer should be pointing out the likelihood/certainty of the aimed change. This can be measured by a concept that has been around for a while:

⚠️ Be conscious about the difference between causation and correlation here: the more leading the indicator is, the less certain you can be about the success of this action. “Outcomes over outputs” generally is a good idea—but not at all cost. It does not help you to pick an outcome indicator just for the sake of it when it leads to a metric of a lagging nature.

⚠️ The nature of the metric highly depends on the context of your company: how your business works, the business cadence, if this metric changes and how often

The most common and risky mistake when using these techniques is to ignore a lack of clarity:

Output key results for Product Discovery are a bit controversial since they are easy to misuse, but can be helpful for teams that struggle with prioritizing discovery work and/or discussing its progress. Metrics like "number of winning prototypes" or "number of opt-ins to user testing" can help make Discovery become a more sustainable practice.

Besides regular (e.g. weekly, bi-weekly) OKR check-ins with the team, make sure to have regular (e.g. quarterly) OKRs retrospectives and make sure that there is a process or person responsible to circle back the learnings to the management team to (potentially) reflect on the way the company measures success altogether.

In order to be able to assess your OKRs continuously and successfully…