Startups Stack Exchange Archive

How do network effect businesses effectively validate product/market fit?

The value of network effect businesses improve with more users. Examples are Facebook, Skype, eBay, Twitter, StackExchange.

These kinds of businesses usually need a minimum viable mass (like Harvard for Facebook) to really validate if they have product market fit. So they build an MVP and also get a MVM to prove they are worthy.

Because of this customer development is critical to validate the product upfront before building it.

What is the most effective way to validate the product before it has a minimum viable mass? How do the other successful companies that don’t luck out on the product do this?

Answer 3010

In my opinion, there is no single correct answer to this question, though I've used a number of different options as an approach to addressing this issue:

Fake it: This approach has been used a lot and has its issues. Core issue is that long-term you can't run a business on a fake network and there's always the potential for backlash legally and to good will. That said, it's a common solution, and as always, it's really up to the business to decide what's best for it.

Simulation: Pretty much the same thing as faking it, though it's clear to the users that the network is being feed/run by the company.

Substitution: This is not the same thing as faking it or simulation. Basically, you find a way to offer substitute for what in the future will be driven by the network. Core issue with this approach is sourcing the substitution and that the substitution is not the same thing as the "real" network.

Proxy: This approach is basically the same thing as substitution, though the intent is to use an existing network real-time as if it was your own. Clearly, there are a number of issues with this, though this has also been done before, and worth noting as an option.

Limited launch: This is often the best approach, that being to very selectively pick an existing network to provide the product to, then if possible, leverage that success to grow beyond the network used to launch the network. This is the approach that Facebook and others has used with great success.

That said, based on my experience, it's possible to do valid customer development without a product that's in final form. Generally speaking, all that's required to validate a network effect based product is that the customer understands that intent and function of the product. Next step is to ask each customer in the first dev-cycle to invite 7-12 of their existing network to take part in the next customer development cycle. If you get enough traction leaping through the networks manually, then that's a very good indicator for the future of the product and its related network. Clearly, having something real to show a customer is better than not, but more often than not you'll learn more by directly engaging customers as soon as possible.

Building on the above, when you're scaling a product knowing the minimum viable network required to onboard new user network clusters is critical. If you're aware that a network has a user that has not been onboarded and another that to your knowledge has no existing connections, you're much more likely to land the user with existing connections than the user with no connections. Further, if you're able to map networks of users before their onboard you might even attempt to find an optimal path through the network to insure you're focusing on the right users to execute a leapfrogging strategy on.

Answer 3012

It depends on the type of network you’re into.

I also think you’re greatly overestimating the importance of reaching a critical mass. Having the latter yields an inflection point in your growth, but the product market fit occurs long before you reach it.

Look for clusters

With some networks, you can have clusters of nodes in the form of users who can make use of the product independently from others.

This is spectacularly illustrated by the fax machine – and telephone before it. Rather than trying to reach out to all businesses and consumers, Xerox reached out for administrations and corporations who have multiple premises – and thus an internal use for the product irrespective of who else might be using it.

Here, achieving product market fit isn’t so much about the market at large as it is about these very special subsets of clients.

(I’d argue that Facebook proceeded similarly: it went for universities before going mainstream, and part of the reason was because alumni networks were very dense and useful in their own right.)

Look for key nodes

Similarly, there can be key nodes in the form of users who are tightly connected to many other users.

I’ve seen this first hand while working into financial extranets. Businesses in finance are all connected to some kind of market data provider, exchange or clearing house. As you might imagine, the key was to focus on these service providers instead of end-users such as banks.

Here again, achieving product market fit isn’t so much about the market at large as it is about these very special subsets of clients.

In the more general case, there is still more to it than merely seeking to acquire a critical mass of users. The key aspect is in fact user retention, and it occurs long before this mass of users is reached.

In the case of Facebook, it was reportedly about having 10 or so active friends. In the case of Twitter, it’s reportedly about following a few users you care about. In the case of a forum or Reddit, it’s about having an answer or some kind of feedback (upvote) soon enough.

The point here is that the product market fit isn’t about user acquisition. It’s about user churn: there’s one or more metric to be identified, and a threshold for each one, beyond which user churn is low enough for the network to grow month to month.

You achieve product market fit when, upon having identified these metrics, and having geared the product towards them, you reach the thresholds often enough that users stay around.

You can try to fake it, or purchase datasets, or create artificial scarcity, etc. to try to bootstrap this all. But nothing will replace the actual marketing work that must go into understanding the characteristics of the network you’re dealing with.

Answer 3060

In the book “The Lean Startup” page 213. Eric Reis says;

Startups occasionally ask me to help them evaluate whether they have achieved product/market fit. It’s easy to answer: if you are asking, you’re not there yet.


All content is licensed under CC BY-SA 3.0.