What are Network effects and Increasing Returns [Part 1 ]
This article (Part 1) is an attempt to explain how we understand increasing returns and network effects at Samaipata. Part 2 is deep-diving into how to create network effects, how to measure them and what the future holds.
Why we focus on Network effects at Samaipata: the power of Increasing Returns
Historically we were living in a world of “diminishing returns”. As explained by W. Brian Arthur back in 1996, in an economy constrained by fixed inputs of capital and labor, businesses that built a dominant position into the marketplace eventually ran into limitations. This world was highly predictable as an equilibrium of prices and market shares was quickly reached.
Yet with the rise of non-tangible, non-rival assets and a shift from this resources-based economy to a knowledge-based economy, it became possible to experience “increasing returns to scale”. This concept refers to the idea that output increases more than proportionally to changes in input levels, as positive feedback mechanisms are triggered: the player ahead gets further ahead, and conversely the player that loses advantage loses further advantages. It thus becomes possible to build defensibility in a sustainable fashion, and businesses can protect long term profits in a competitive environment, with these increasing returns powering winner-takes-all markets.
There are different ways to generate increasing returns:
- Economies of scale: increased input levels lead to a non linear decline in unit costs.
- Customer “Groove-in”: switching costs materialise due to investment in product usage and/or brand equity.
- Network effects (Nfx).
Generally great companies reinforce network effects by other sources of defensibilities. Doordash is a good example of this “reinforcement effect”, which powering a strong growth “flywheel” (See below an extract from their S1). That being said, at Samaipata, we are passionate about network effects because we believe they are one of the most powerful drivers of defensibility. They are responsible for c. 70% of value creation since the advent of the internet back in 1994 according to US fund Nfx.
A key question we are often asked is: are there successful companies out there that are not displaying Nfx? We think that it could be possible to dig moats in non-networked products: some businesses’ value propositions are not directly tied to the number of people using the product (e.g. in Biotech). But in this case, we would need to see huge scale, strong brand equity or switching costs to be interested, and those businesses do not typically fit our investment criteria.
Definitions & considerations about Network effects
The term “network effects” was coined by AT&T in their 1908 Annual Accounts: “a telephone — without a connection at the other end of the line — is not even a toy or a scientific instrument. It is one of the most useless things in the world. Its value depends on the connection with the other telephone — and increases with the number of connections.”
We see network effects occurring when i) a company’s product or service becomes more valuable as more users join the network and/or when ii) usage/engagement between users in the network increases. These network effects are typically enabled by many-to-many connections, which is the core of our investment thesis.
Network effects generally kick in once a certain “critical mass” has been reached: before this point, the product remains quite vulnerable and may not have much value to users. And conversely, from this point onwards, the value produced by the network exceeds the value of the product itself and of competing products. The famous “chicken or egg” problem refers to the challenge of getting to this Minimum Viable Critical Mass (as coined by Sean Ellis) to trigger a positive feedback loop.
Yet not all Nfx are created equal: Nfx are very complex and different in nature. And companies can display different types of Nfx at the same time. Our General Partner, José, has written an entire article on Nfx classification where he comes back to the difference between:
- Positive linear vs. asymptotic vs. negative Nfx
- Direct vs. indirect Nfx
- One-sided vs. multi-sided Nfx
- Local vs. global Nfx
Ideally positive, non-asymptotic, multi-sided & global Nfx are the most powerful type of Nfx we’ve seen in the past. Yet we are also super interested in data Nfx, that occur when the product & services’ value increases with more data. That being said, we’re conscious that data Nfx tend to be weaker than many people often want to believe and might take time to materialise. We will publish more about this in the coming weeks.
Additionally, one of the most obvious and well-known types of Nfx are “marketplace Nfx”: in the transaction, there are two (or more) sides involved with very different interests that directly benefit each other (e.g. supply & demand on Craiglist). But we are also strong believers of “platform Nfx”: the products & services created and sold by the suppliers are a function of the platform, not independent of it (e.g. iOS developers vs. iPhone users).
How are Network effects different from viral effects and scale effects?
Nfx vs. Viral effects
Viral effects are about growth of new users: they happen when you get your existing customers to get you more new customers, ideally for free. Conversely, Nfx are about defensibility of your model: when every user of your product adds incremental value to all the other users so that it becomes difficult for users to find any alternative product which gives them as much value.
You can have one without the other: for instance, Sequoia-backed QuizUp (which raised $27M and died) had high virality but no Nfx; alternatively, some B2B marketplaces have strong indirect Nfx but are not growing virally.
Yet they are generally complementary as virality helps you achieve network effects by growing the network more quickly.
Nfx vs. Scale effects
Economies of scale happen when there is a reduction in marginal costs due to an increase in production (e.g. Apple amortising Capex on an increased number of iPhones sold). It’s about costs! Conversely, Nfx (also known as “demand side” economies of scale) happen when production has a greater value for the same marginal cost of production. It’s about value! And value should increase exponentially while costs decrease linearly.
Yet there are both ways to generate increasing returns. And Nfx typically magnify a company’s scale advantage and vice versa…
In Part 2, we’re addressing how to create network effects, how to measure them and what the future holds for them.
And as always, if you’re a European platform founder looking for Seed funding, please send us your deck here or subscribe to our Monthly Founders Kit here!
Samaipata is an early stage founders’ fund investing in digital platforms displaying increasing returns at scale, across Europe.
