Five things, not a document
- What happens today, as numbered steps — even rough ones
- How often, and who does it
- What goes wrong and what it costs — errors are usually where the business case lives
- Which systems are involved, named
- A budget range you would be comfortable with
Half a page covering those five gets you a better answer than four pages covering anything else.
Why we ask for the budget
Withholding it produces proposals scoped at the wrong scale. A range — “we expect £15,000–£30,000 and want to understand what each end buys” — gets you comparable answers from anyone you ask.
It does not mean you will pay all of it. It means nobody wastes a fortnight designing something you were never going to approve.
What helps more than prose
- A screenshot of the spreadsheet you currently use
- A sample of the emails or documents that arrive
- A screen recording of someone doing the task, however rough
- Last month's numbers — volumes, errors, hours
Fifteen minutes of these is worth more than a written specification, because it shows what actually happens rather than what people believe happens.
What not to bother with
A technical solution design, unless you have one you are committed to. A long company background. A feature list without the process behind it.
We would rather understand the problem and propose the approach than price a solution somebody else specified.
What you get back
A reply without being left waiting, either with two or three questions or with a suggested time to talk. Then a rough cost range, a rough timeline, the main risks, and a clear next step.
Including, sometimes, that there is no project here and why.