Network Effects vs Network Economies: The Materiality Test
Almost everything has some network effect; only ask whether it's large enough to change your margins.
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 3
- Confidence
- 92%
Helmer draws a sharp line between a network effect (present in almost any platform business, often trivially) and a network economy (a genuine power). The distinguishing question is materiality: does the effect create a value benefit large enough to support a price delta that gives you materially different margins into the future? If it's worth a penny to the bottom line it's just an effect; if it's worth a billion dollars it's a power. The same materiality lens applies to 'flywheels' and to claimed data scale economies.
Origin
Hamilton Helmer, '7 Powers' — his insistence that a 'type of power' must clear a significance/materiality hurdle, distinguishing his 'network economies' from the looser popular term 'network effects'.
Core principles
- 01Almost any platform business has some modest network effect
- 02A network effect only becomes a network economy when its impact on margins is material
- 03Materiality means it engenders a price delta big enough to move future margins, not a penny
- 04The same test debunks most 'flywheel' and 'data scale economy' claims
- 05Data scale economies are real but rare because the cost curve flattens once rivals are within shouting distance
How to run it
- 1
Assume the effect exists
Don't argue about whether a network effect or flywheel is present — in a platform business it almost always is. That's the wrong question.
Pro tip Turn almost anywhere and you'll find some modest network effect; its mere existence proves nothing.
- 2
Measure the materiality
Ask whether the benefit is large enough to engender a price delta significant enough to give materially different margins into the future. Estimate the actual dollar impact on the bottom line.
Pro tip Frame it as penny-vs-billion: quantify roughly whether the effect moves margins meaningfully.
Watch out A real-but-small effect (Helmer's read on Uber/Lyft) is not a power — the advantage isn't material.
- 3
Apply the same lens to flywheels and data
Re-run the materiality test on any 'flywheel' or 'we get scale economies from data' claim. For data, check whether existing competitors are already at enough scale that the per-unit cost curve has flattened — if so the advantage is immaterial.
Pro tip Most fixed-cost scale economies flatten: the percentage cost advantage shrinks as everyone gets big.
Watch out Data scale economies fail most often not because they don't exist but because rivals are within shouting distance on cost.
In the wild
Both ride-hailing firms likely have some network effects, but Helmer argues they don't have network economies — they must keep spending heavily to stay ahead, so the advantage isn't material. He attributes Uber's win instead to a well-played war of attrition on top of modest, geographically-specific scale economies.
→ A case where visible network effects never translated into a durable, margin-moving power.
Companies claim scale economies from data, but if incumbents already operate at scales large enough that their cost-per-unit is within shouting distance of each other, the curve has flattened and the cost advantage is small.
→ A common claimed power that fails the materiality test.
Common mistakes
Pitching 'we have a flywheel' as a power
Flywheels and network effects exist almost everywhere; the ones that materially tilt returns are rare, so their existence alone is not evidence of a barrier.
Assuming data automatically means scale economies
It's possible but rare — competitors are often already at enough scale that the fixed-cost advantage curve has flattened and the per-unit difference is immaterial.
Is it for you?
Best for
Founders and investors stress-testing a 'network effect' or 'flywheel' or 'data moat' pitch before treating it as durable advantage.
Not ideal for
Situations with no comparable competitors or no cost/price data, where materiality can't yet be estimated.
From the transcript
“there there are lots of things that I would say have Network effects but not Network economies”
“the difference is materiality”
“if it's a penny to your bottom line that's one thing if it's if it's a billion dollars or something else”
“we sort of laugh whenever we say we hear somebody say they have a flywheel which kind of gives you the idea of network economies…”
“the curve flattens in other words which is typical of any because the most common scale economy is You' got a big fixed cost”
From the episode
Business strategy with Hamilton Helmer (author of 7 Powers)