The most expensive AI mistakes are made at the start, when a vague idea goes straight into a build. Scoping is the cheap insurance against that: a short, deliberate effort to define the problem, check whether the data exists, agree what success looks like, and decide honestly where AI belongs - and where it does not.
Here is a practical way to scope an AI software project so the build that follows is the right one.
Start with the problem, not the technology
Write down the problem in terms of the outcome you want and the constraints you are under, before mentioning any model or tool. What does a good result look like? Who acts on it? What happens today, and what is wrong with that? A scope that starts from the problem tends to build the right thing; one that starts from "we should use AI" tends to build a demo.
Check whether your data is ready
AI and analytics are only as good as the data behind them. Before committing to a build, establish what data exists, where it lives, how clean it is, and whether you are allowed to use it for this purpose. "Data readiness" is often the real work - and discovering a gap now is far cheaper than discovering it mid-build.
- Does the data you need actually exist, and can you access it?
- Is it clean and consistent enough to rely on?
- Are you permitted to use it for this purpose?
- Where AI is involved, what data would be sent to which provider?
Decide where AI belongs - and where it does not
Separate the parts of the workflow that benefit from probabilistic help from the parts that must stay deterministic. Not every step needs a model, and the honest answer is sometimes that a simpler rules engine - or no AI at all - is the right tool. Making that call deliberately is part of good scoping, not a failure of it.
Agree success criteria and a first phase
Define what "good" means in measurable terms, and agree acceptance criteria before the build begins. Then scope a clear first phase with a known budget rather than an open-ended commitment, so you can see working software early and decide, with evidence, whether to continue. A scope that can be delivered and judged in phases is a scope you can trust.
