I've seen this model used to develop domain specific decision support software used to help customer optimise costs. Required a few person years of R&D to hit p

I've seen this model used to develop domain specific decision support software used to help customer optimise costs. Required a few person years of R&D to hit proof of concept & get a contract signed, then contract was large enough to fund a small business & grow headcount for several years. One potential trap is over-fitting the software/product to what the first big customer wants, or getting carried away spending too much time doing random bits of customer specific consultingware, and not being able to factor out a product that can be generalised and sold to other customers with somewhat different requirements.