Building Software Without an SRS? Here's Why Projects Go Wrong

The customer management example is a good one, because follow up reminders and employee permissions are exactly the things two people picture differently until someone writes them down.

One thing I would add from experience: the part of a requirements document that pays off most is the non-functional section. Clients rarely forget a feature, but response times, data retention and who can see what are usually left implied, and those are the decisions that cause rework late in a project.

I would also push for every requirement to end in a check someone can run. "Users can manage customers" is hard to accept or reject. "A sales rep can see only their own customers and a manager can see the whole team" is easy.

At Ascendra Ventures we scope work as one defined problem with a fixed price, and that only works because the acceptance checks are agreed first. Do you find smaller clients are willing to spend time on an SRS, or do they want to start building straight away?