Learn While Growth Is Working
Decode current growth before momentum hides its causes and reaches a plateau
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 99%
Learn While Growth Is Working reverses the instinct to investigate only when metrics deteriorate. Growth will eventually decelerate mathematically, so teams should use periods of momentum to identify the user segment, core use case, distribution channels, seeding strategy, and growth loops responsible for success. They test whether these mechanisms are causal and repeatable rather than assuming the visible trajectory will persist. This understanding gives the company levers it can strengthen when organic momentum fades and informs which people it should hire. The framework also guards against accidental product-market fit: a company can possess fit without understanding why, leaving it unable to reproduce or extend the result. The output is a documented causal model and a prepared next phase of growth.
Origin
Extracted from Marketing Against The Grain through Anu Atluru's warning that successful periods are often the hardest times to learn.
Core principles
- 01Every growth curve eventually decelerates
- 02Success can conceal weak causal understanding
- 03Current momentum creates time to prepare the next phase
- 04Core use cases and growth loops should be understood before a plateau
- 05Product-market fit can exist without the team knowing why
How to run it
- 1
Name the Growth Cohort
Identify which users, use cases, channels, and behaviors contribute disproportionately to current growth. Avoid treating aggregate growth as one homogeneous phenomenon.
Pro tip Compare retained cohorts rather than acquisition totals alone.
Watch out A large surge from one transient channel can look like broad product-market fit.
- 2
Find the Core Use Case
Determine the recurring job or experience that causes valuable users to return and recommend the product. Build conviction through behavior and conversations.
Pro tip Ask what successful users would replace the product with if it vanished.
Watch out Do not confuse the most visible feature with the core retained value.
- 3
Map the Growth Loops
Trace how one user's activity can produce another user, more content, stronger retention, or additional value. Mark where the loop depends on manual intervention or temporary conditions.
Pro tip Express each loop as an observable sequence with measurable transitions.
Watch out A channel that requires continuously increasing spend is not automatically a self-reinforcing loop.
- 4
Test Causality During Momentum
Run focused experiments and qualitative checks to determine which mechanisms truly drive the observed result. Use the healthy growth period as learning runway.
Pro tip Preserve a record of major changes and cohort responses.
Watch out Do not stop investigating merely because headline metrics are rising.
- 5
Prepare the Next Phase
Choose the loop to deepen and a small set of new levers to test before growth slows. Align hiring and product investment with the resulting conviction.
Pro tip Start the next-phase work while the current engine still funds experimentation.
Watch out Waiting for a plateau forces the team to learn under pressure and with less runway.
In the wild
A social product grows rapidly through recommendation and network seeding. Rather than only hiring to handle demand, the team identifies the retained user archetype, instruments invites and feature use, and maps which behaviors generate valuable new users.
→ When growth decelerates, the team already knows which loops and use cases to strengthen.
Common mistakes
Celebrating Without Diagnosing
Strong growth can tempt a team to pursue the next shiny object without understanding the engine already working.
Waiting for the Plateau
Beginning causal analysis only after growth stalls wastes the period when the company has the most momentum and optionality.
Is it for you?
Best for
It is best for teams experiencing rapid adoption, strong word-of-mouth, or an early product-market-fit inflection.
Not ideal for
It is not ideal for teams with no meaningful usage signal yet, where basic problem and user discovery must come first.
From the transcript
“sometimes when things are working the best is actually when it's the hardest to learn.”
“by the time you get there, you want to know what the levers of growth are and what your growth loops are in your product…”
“it's possible that you have product market fit without really understanding why.”
From the episode
Community-Led Growth vs. Product-Led Growth with Anu Atluru
Anu Atluru