...and then they rejected it. Lovely idea to spend years making software that only one business can use.

I'm not saying that the code these guys made can't be used for other projects, there are probably lots of opportunities...but there should be a lesson here about considering your market before making any commitments.

Unfortunately it's the way it works in some areas. Just like with any other startup: you work and work to get something done and then hope your clients are willing to pay for it. Except in some markets the entire user base is a handful of companies with really deep pockets. I guess you should always try to get feedback as soon as possible but it's not always easy depending on who you're dealing with.