Fall in Love With the Problem
Keep the problem stable while experiments replace weak ideas
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 99%
Anchor the venture to a persistent customer problem, not to the first solution imagined for it. An idea is a temporary hypothesis and will usually reveal flaws when tested. The problem can remain valuable while individual ideas are discarded. Define it independently of product features, expose a proposed solution to real customers, and use each failure to make the problem model richer and more precise. Then generate another solution from that improved understanding. This orientation makes pivots coherent rather than random: the implementation changes while the learning compounds around a stable need. It also protects founders from defending an attractive idea long after evidence has rejected it. Commitment belongs to understanding and solving the problem; flexibility belongs to every proposed way of doing so.
Origin
Randolph rejects simplified eureka stories about Airbnb, Uber, and Netflix. He describes innovation as a random, meandering path in which failed experiments change ideas while making the underlying problem richer.
Core principles
- 01Ideas are temporary hypotheses while worthwhile problems persist
- 02Failed experiments should deepen understanding of the problem
- 03Attachment to one solution blocks useful pivots
- 04A richer problem model produces better subsequent tests
How to run it
- 1
Name the problem
Describe the customer's persistent struggle without mentioning your product, feature, or preferred solution.
Pro tip Use language that would remain true if the current idea disappeared.
Watch out A solution disguised as a problem will preserve attachment.
- 2
Frame the idea as temporary
State the current solution as one hypothesis about how the problem might be solved.
- 3
Test with customers
Expose the hypothesis to real behaviour and gather evidence about both the idea and the problem.
Pro tip Listen especially for evidence that changes your understanding of the need.
- 4
Enrich the problem
Record what the failed or surprising result reveals about context, urgency, constraints, or desired outcomes.
Watch out Do not preserve the idea by explaining away contrary evidence.
- 5
Replace the solution
Discard or revise the idea while keeping the enriched problem as the anchor for the next experiment.
Pro tip Ask what solution becomes plausible only after the latest learning.
In the wild
A team trying to improve movie rental discovers that one delivery and pricing idea does not work. Instead of protecting that implementation, it keeps investigating the customer's difficulty finding and accessing enjoyable stories, then tests a different model informed by what failed.
→ The solution evolves while learning about the customer problem compounds.
Common mistakes
Defending the first solution
Treating the idea as the venture's identity makes contrary evidence feel like a threat. Keep identity attached to the problem instead.
Pivoting without an anchor
Changing both problem and solution after every result produces random motion. Preserve the validated problem while replacing weak hypotheses.
Is it for you?
Best for
It is best for early-stage teams exploring a meaningful problem through multiple uncertain solutions.
Not ideal for
It is not ideal when there is no evidence that the chosen problem exists or matters to the intended customer.
From the transcript
“fall in love with the problem don't fall in love with the idea”
“if you have the right problem the problem stays with you forever”
“each failed experiment makes the problem richer and more complex”
From the episode
Marc Randolph: How Netflix Disrupted The Entertainment Industry
Marc Randolph