A network effect exists when the value of a product, service or platform changes as participation in the network changes.
The simplest example is communication.
A telephone is more useful when more people you want to contact also have one.
Platforms can produce a different form of network effect. More sellers may make a marketplace more useful to buyers, while more buyers make it more attractive to sellers.
Network effects can create powerful growth loops, but they are frequently claimed where they do not actually exist.
Table of Contents
What are direct and cross side network effects?
A direct network effect occurs when more participants of the same type increase value for one another.
Messaging and social networks are common examples.
A cross side network effect occurs when growth on one side increases value for another side.
More drivers can improve availability for riders. More riders can make the platform more attractive to drivers.
The strength and direction of the effect matter.
More participants do not always make a system better. Too many sellers can make discovery harder. Too many users can create congestion. More competitors can make a platform less attractive to individual complementors.
So network effects should be tested, not assumed.
How are network effects different from scale?
Scale means an organisation becomes larger.
A network effect means that participation changes the value experienced by other participants.
Those are different mechanisms.
A manufacturer may double production and reduce unit costs. That is an economy of scale.
A marketplace may double its number of relevant sellers and make it easier for customers to find what they want. That may be a network effect.
The distinction matters because strategies differ.
Economies of scale often reward efficient expansion.
Network effects require attention to participant mix, interaction quality and whether adding one group actually creates more value for another.
Do all ecosystems have network effects?
No.
Ecosystems are built around interdependence and complementarity, but network effects are not required.
A pharmaceutical innovation ecosystem can involve several specialised organisations whose contributions must fit together without becoming more valuable simply because more participants join.
An industrial solution ecosystem may need exactly the right set of partners rather than the largest possible network.
This is why the phrase we have network effects deserves scrutiny.
Ask:
Who becomes better off when another participant joins?
How much better off?
Does the effect continue as the network grows?
If those questions cannot be answered, growth may simply be growth.
Why can network effects change competitive power?
Strong network effects can make an ecosystem harder to challenge.
A new entrant may need customers to attract complementors and complementors to attract customers. The incumbent already has both.
But this does not guarantee a winner takes all market.
Participants may use several platforms.
Customer needs may differ.
Switching may be easy.
Local markets may remain separate.
Congestion or falling partner economics may weaken the benefit of scale.
The result depends on how strong the network effects are and how difficult it is for participants to move or use alternatives.
How should network effects be measured?
Look for evidence that participation changes behaviour or value elsewhere in the system.
For example:
Do more relevant sellers improve buyer conversion or retention?
Do more users make complementors more willing to invest?
Does greater local density reduce waiting time?
Does more variety increase usage?
Then look for limits.
At what point does choice become clutter?
Does partner profitability fall as competition rises?
Do users start to split activity across several ecosystems?
A useful network effect is not a story about size.
It is a measurable relationship between participation and value.
Related concepts
Complementarity
Multihoming
Ecosystem Growth
Ecosystem Tipping
Ecosystem Advantage
Ecosystem Value Creation