Acquiring unique, high-quality data, especially from the 'hard side' of the marketplace (e.g., candidates and experts), is paramount as proprietary data differentiates an AI product

It is important to use "Kindle strategies" for early growth, which are whatever methods can acquire the next few hundred users to validate product changes. These strategies may not be scalable long-term but help in sequencing towards more sustainable growth

Building new networks is particularly challenging now because AI hasn't opened up "new land" like mobile did. To create a new network, it often means taking market share from existing ones.

Consumer network companies face a "three-body problem": they need the right target audience, the right network size, and the right value proposition to succeed.... See more

Historically, many founders found a product that worked and then sought to grow it. However, a product that truly works should already be growing, meaning it has a built-in mechanism for growth

Additionally, product-channel fit remains essential. Products must be designed to tap into specific channels, as channels cannot be molded to fit a product. This means considering how the product will naturally acquire users through existing channel

The fastest-growing products are those that build growth loops—compounding systems that turn one user into another. Building these loops requires a deep understanding of users, focusing on why they do something rather than just what is desired of them