| The textbook approach is: 1. Find product market fit 2. Get dominant market share 3. Work on the next cycle - the core business will go stale 4. Everything else in support - legal, finance, HR Look for the Step 3. People think of experimental teams as the ones that build Gmail or Android, but they're more the ones building Gemini. It also happens quite early on, maybe as early as Series B companies. You might want to avoid 'toy' teams. Gmail was probably the most successful toy project, but they're very rare. One example is a product that has lots of TPV/GMV - payment gateways, marketplaces, listing sites. This usually doesn't translate to revenue/profit, but they're in a good position to sell things that are profitable. Another is fixing problems in scale. Like FB was built on PHP which wasn't enough, so they ended up building Hack, React, a bunch of mobile stuff. Some of the work is to build features requested by the core. An elliptical example here might be if you were Rolls-Royce and needed a specialized exhaust system, the specialist team would build it and then try to mass produce or sell that exhaust part to other companies. There's going to be a lot of failed products and that's okay. The fail rate is usually higher than the core product, but lower than startups, so it's ideal for former startup folks. |