Most scoping exercises take a month and still produce wrong estimates. Here is our one-week process.
A discovery that takes three months is not more accurate than one that takes one week. It is just slower — and by the time it finishes, requirements have already changed. Our scoping process is designed to get to a credible estimate in five working days without cutting corners on the parts that actually matter.
We run a structured requirements workshop with every stakeholder in the room. The output is a prioritised list of user stories in MoSCoW format: Must, Should, Could, Won't. We are ruthless about the Won't column — features that are not in scope for v1 go there immediately, no matter how attractive they sound.
We sketch the data model before we think about screens. Every meaningful noun in the user stories becomes a candidate entity. Relationships between entities reveal the real complexity. A simple-sounding feature often requires three tables and a complex permission model when you look at the data side.
For any integration or technology we have not used before, we run a time-boxed spike — typically four hours. Spikes answer specific questions: can we read from this third-party API in real time or do we need to cache? Does this payment gateway support split settlements? Spikes kill assumptions.
We estimate at the story level using three-point estimation (optimistic, realistic, pessimistic), aggregate, and apply a 20% risk buffer. Then we sense-check the number against similar past projects. If the number is materially different from past projects, we find out why before presenting it to the client.
Discovery call within one business day. No commitment.
Product launches, case studies, and IT-services tips. No spam.