Discover more from The Product-Led Geek
Minute Monday #3: The 5 Key Questions of an Impact & Learnings Review
Avoid blind success by seeking to learn why you won as voraciously as you seek to learn why you failed.
In my view, the Impact and Learnings (I&L) review is the single most important ceremony for a growth team.
It’s the opportunity to reflect as a team on the new knowledge you’ve acquired and the implications of that on where you go next.
The learnings should be the star of the show, with the impact there for context.
In this post, I want to give you a rundown on the 5 most important questions that should be asked and answered in an I&L review.
1. What did we do?
Get everyone on the same page by sharing how you arrived at the learning.
Was this an experiment? Qualitative research? Quantitative exploration?
Describe what you did in enough detail to provide context for alignment.
2. What did we believe was going to happen?
Provide detail on what you expected to see/observe/happen.
Why did you expect those things?
3. What actually happened?
Given we expect maybe 1 in 5 experiments to validate our hypotheses, it’s typical that what you expected to happen does not.
Insight comes from analysis and conversation around the observed outcome, including any delta from the hypothesised outcome.
4. What did we learn?
Describe the learning itself
Have you established causal relationships?
Do you understand why you’ve observed something?
What were the effects on secondary and health metrics, and what does that tell you?
Where else could you apply what you learnt? In what other contexts might these learnings be relevant?
Have the learnings led to other hypotheses?
5. What do we intend to do next?
As a result of your learnings, what do you intend to do next?
Some examples (not exhaustive and not all mutually exclusive) are:
Implement learnings at scale to create <xyz> impact
Experiment further to learn <xyz> or validate hypothesis <xyz>
Analyse event data to understand <xyz> better
Conduct user interviews to understand <xyz> better
Conduct the same experiment in a similar area <xyz>
Deprioritize - we’ve learnt <xyz> and decided not to move forward in this direction.
Thanks for reading The Product-Led Geek! Subscribe for free to receive new posts and support my work.