01 / PLANNING
Start with one business outcome.
“Build a management system” leaves too much open. A more useful brief describes a workflow: a customer request becomes a job, the job is assigned, completion is recorded and the invoice is reconciled. List the people involved and the information each person needs.
Separate the first release from later improvements. For example, a dispatch team might need job assignment and delivery status first, with route optimisation considered later. A smaller first release still needs the controls required to operate safely.
02 / PLANNING
Look beyond the number of screens.
Integrations, permissions, approval rules, reporting and data migration can shape the effort as much as the visible interface. A payment integration, for example, needs a plan for unmatched transactions, retries and reconciliation as well as the initial payment.
Record what exists today: spreadsheets, software subscriptions, available API access and the quality of the records to be migrated. Unknown dependencies should be investigated or identified as assumptions in a proposal.
03 / PLANNING
Compare the same scope across proposals.
Check which workflows, integrations and reports are included. Ask how acceptance will be demonstrated, who provides data and feedback, and how changes to scope are handled. A price is difficult to compare without those details.
Discuss hosting, third-party subscriptions, maintenance, training and handover separately. Clarify ownership and source-code access in the agreement. A lower initial build cost does not describe the full cost of running the system.
04 / PLANNING
Use your current workload as a baseline.
Estimate how often a task happens and the time it takes. If a task happens 20 times a week and takes 15 minutes each time, that is five hours of work per week. This is a workload estimate, not a promise that software will remove all five hours.
Include the effort to introduce the system: cleaning data, reviewing designs, training staff and changing routines. Use the baseline to decide whether the proposed improvement is worthwhile, then agree a scope and budget with the delivery team.
