Skip to content
Graph of Plan

Breadth Beats Depth For Market Learning

Shipping 4 products at 70% quality beats shipping 1 product at 95% quality for learning what the market wants.

The Assumption

The plan includes 4 products: SmartBoxes, Murphy, Nomos Cloud, P4gent. This is a portfolio bet—we don’t know which will win, so we’re testing multiple.

But is breadth actually better? Counter-arguments:

  • Infrastructure rewards depth (reliability, trust, features)
  • Context-switching destroys productivity
  • “Jack of all trades, master of none”
  • 70% quality might not cross the threshold for any product

Maybe 1 product at 95% beats 4 at 70%.

Evidence

Supporting signals:

  • Startups are search processes—more experiments, more learning
  • Early-stage products don’t need to be perfect
  • Market feedback is more valuable than polish
  • Successful founders often explored multiple ideas

Counter-signals:

  • Infrastructure has higher quality bar than apps
  • Users don’t forgive buggy developer tools
  • Splitting attention means nothing gets done well
  • Compounding: depth in one area builds moat

What Would Prove This Wrong

  • All 4 products fail due to insufficient quality
  • Context-switching prevents any product reaching the bar
  • Market clearly rewards depth (competitors with 1 product win)
  • No meaningful learnings transfer between products

Impact If Wrong

If depth beats breadth:

  • Narrow focus to 1-2 products immediately
  • Accept slower market learning
  • Build deeper moat in chosen area
  • Change sequencing strategy

Testing Plan

Signal quality:

  • Track learnings per product (are they distinct?)
  • Track cross-product learnings (do insights transfer?)
  • Compare signal quality: 1 deep product vs. 4 shallow?

Quality bar:

  • Are any products crossing the “good enough” threshold?
  • User feedback: is quality a blocker?

Review: Quarterly assessment of portfolio vs. focus strategy

Depends on:

Affects:

Assumption

Shipping 4 products at 70% quality beats shipping 1 product at 95% quality for learning what the market wants.

Depends On

This assumption only matters if these are true:

How To Test

Track learnings per product. Compare signal quality from multiple bets vs. single bet.

Validation Criteria

This assumption is validated if:

  • Multiple products generate distinct market signals
  • Portfolio approach identifies winning product faster
  • Learnings from one product improve others

Invalidation Criteria

This assumption is invalidated if:

  • All products fail due to insufficient quality
  • Context-switching prevents any product reaching quality bar
  • Market rewards depth over breadth in infrastructure

Dependent Products

If this assumption is wrong, these products are affected: