top of page

Cold start problem and network effects

Ever left your car in the cold for days and had a problem starting it up again? It goes something like the engine not turning over, no combustion, no spark and eventually when you least expect it something happens and the engine finally fires up and you scramble to drive out. The cold start problem in product management is similar where a product is struggling to fire up its engine and get going to scale its user base exponentially. The cold-start problem leads to anxiety, self-doubt and second guessing. This situation is the hardest problem start-ups and PMs face in scaling up their early-stage concept / MVP/ MLP to a flourishing eco-system, where the product achieves commercial success and scale of adoption.

One way to solve the cold-start problem is to create a “network product” where each user (akin to a node) will add value to the platform by introducing other participants (thereby branching out to create many more nodes) and by using the product to connect further with their own network of customers and partners, in theory each node creates its own mini club. By harnessing the power of the network every dollar invested into the platform should net a higher return multiplier and the hope is that the inertia transforms into a sustainable momentum creating a flywheel effect.

The traditional approach to product scaling has been by to grow organically. This is a linear growth model where users are acquired through a direct relationship, at best a referral model where one user refers to another user to the service. This approach is still used but it doesn't deliver exponential sky rocketing growth which start-ups need to increase their valuations and delight the venture capitalists. The path to getting a network to thrive is helluva lot of trial and error with a good amount of financial backing to continue experimenting and making the best decisions based on the available information at that point in time.

There is no magic pill which can be used to replicate success. Many of the growth stories you hear about have not happened overnight. Zoom which has been the darling of the pandemic with its video conferencing toolkit was not an instant hit, the product has been in the making for nine years. Think of the amount of labour, sweat, tears and dooms day scenarios the team has gone through. Sometimes we also need a some luck to come our way, preparation meets expectations and when the timing is right we should be ready to seize the moment. In the zoom example when the pandemic struck companies were caught off guard and many opted to use Zoom for business communications, this led to a snowball effect where other businesses signed up and eventually people used it to stay in touch with family, friends and wider community. To be clear a network product is not a viral product.

The cold start problem and the hypothesis on how to solve it by creating a network product differs dramatically between B2C and a B2B. This is because in B2C the consumer is just one individual who is making the decision and determining if the product is fit for purpose, whereas in the B2B construct there are many decision makers who have their own opinions and the end user doesn't have the sole authority to decide if they want the product. The other reason why the cold start problem is easier to solve for in B2C is the abundance of data and research which makes it possible to make more informed decisions which increase the chance of success.

As PMs collect and analyse data to decide what’s next they should proactively look for growth trends and nudge on whatever is working to improve the trajectory and give it the best chance to reach the tipping point where the laws of nature will take it to explosive growth. The tipping point occurs when the product begins the dominate within the addressable market - this is often confused with the point at which product moves from a relatively modest growth track to high growth. Given that some markets are big enough for competing products to co-exist it is not necessary that a true convergence needs to occur where the product needs to dominate. As an example look at the market shares for music streaming (Spotify, Tidal, Apple, YouTube) and subscription video on demand (Netflix, Amazon, YouTube, Apple, HBO, Disney).

In a basic network there is a starting point and an end point. For the network to prosper and get the maximum throughput aka “transfer of value” both the nodes need to get what they want from this relationship. When you provide a product on the network then expect your counter-party to seek value and demand value from you not just in what is being provided at the present point in time but to also for the future. The product needs to evolve for the network to be sticky. With instant access to competing digital platforms your user is just clicks away from switching. This is why companies and product managers need to understand in a network strategy the product is never completed, the product is just solving a problem at that point in time. Netflix started with Standard Definition videos and progressively moved to High Definition and now Ultra High Definition - along the way with technology advances in internet bandwidth 8K is a real possibility for the mass market. In summary the decision to stick with a network is based on the present value of the future value of the network. Early adopters for this reason are willing to forgo the shortcomings of the network as they believe in the vision “the why” - the network will succeed.

Both sides of the network have their own power. The dimension of the power balance dictates the strategy and path to scaling up. It’s not always a one sided game where the service provider has to worry about influencing the other party. The other party exerts their power by positioning their response as “what’s in it for me”. This friction of getting counterparties to participate needs to be overcome as they bring the real value to the network - every relationship has a hard side which needs to be looked at, start with the elephant in the room. Direct network effects occur when the value of a product or service to your user increases exponentially with the number of the other users using the same product or service. Indirect network effects come into play when the value of your product or service increases due to complementary products or services that add to your platform/business. Create multiple dimensions of networks for your products to thrive. Allow each network node to grow to create an effect of individual scaling, let each note be their own private club of services. This allows the network to thrive faster as each note is not equally weighted in its value. The holy grail in creating a true Network Effect is to end up with a flywheel of economic growth but before all that remember to keep placing many micro bets on the product and design to widen the circles of adoption. Stay on top of the user feedback and commit to developing the roadmap rapidly.

Recent Posts

See All


bottom of page