Lean Startup is not about building trivial things and not hiring sales people. It's about testing business assumptions. And it can work for any type of business

Lean Startup is not about building trivial things and not hiring sales people. It's about testing business assumptions. And it can work for any type of business, even non-technical businesses and companies with ambitious goals. The "Lean" in Lean Startup does not mean "Don't Spend Money". It's a reference to Lean Manufacturing, and it's about not wasting resources. Waste in manufacturing come from things like excess inventory, needless movement of people and parts, and excessive defects. In software it's things like developing features users don't want, or onerous release processes. A lot of people think Lean Startup is all about the Minimum Viable Product, and if the project is too ambitious for a little MVP you can't use Lean Startup. But that's not true. The MVP is about testing business assumptions; your leaps of faith. Here is an example of an ambitious project, the kind that will take years to develop: Lets say you want to build something like Google Spanner/F1. A globally consistent database with rich query capabilities and reasonably low latency. Do you release minimal versions to test if people want it? No, you can't. Minimal versions exist. We have globally inconsistent, or rich query, or low latency databases already. You have to do the market research and decide that if you build it there will be a market. Your testing of business assumptions comes later, after you develop it. For example, do you sell it as a hardware appliance? I say that's probably a bad idea, the companies using that sort of database aren't big on proprietary hardware. What about packaging it as proprietary software? Maybe, plenty of companies buy proprietary databases. But banks aren't early adopters (except Deutsche Bank), and you'll miss "the next Facebook" during their early development/growth phase. What about releasing it as open source and making money off support and services? I think you'll get the startups using it, and once they grow a little I think enough will pay to make the business profitable. But those are all my business assumptions. Those are the things that need testing. Maybe hardware appliances are what the market really wants. But if they aren't (and you assumed they were) you would be wasting money developing hardware. Maybe it's not tech companies that want this, but actually banks that would love a hardware appliance. Or scientific research institutions with big budgets that are happy to use proprietary software. You can use Lean Startup methodologies in any company and in any part of the development phase.