Most bad estimates are not wrong about the work they contain. They are wrong about the work they forgot, the effort they buried, and the questions they never resolved.
Run this as a fifteen-minute review with someone who did not build the estimate. The value comes from a second reader, not from the list.
How to use this
- Review before the estimate is shown to a client, while the shape can still change.
- Any item you cannot answer is either a question to resolve, an assumption to state, or scope to exclude — never padding.
- Repeat the final section after delivery. Comparing estimated to actual effort is what turns your estimating into an asset instead of a habit.
Coverage
- Every requirement captured in discovery maps to at least one estimate line, or to a stated exclusion.
- Every estimate line maps back to a requirement someone asked for.
- Discovery, analysis, and solution design effort is included, not absorbed.
- Testing effort is included: functional, integration, performance, and client acceptance support.
- Data migration, cleansing, and reconciliation are estimated separately from build.
- Deployment, cutover, and any parallel-run support are included.
- Documentation, training, and handover to the client's own team are included.
- Warranty, hypercare, or post-go-live support is either included or explicitly excluded.
Drivers and structure
- The quantities behind each line come from the client, not from an assumption you made silently.
- The two or three drivers that explain most of the effort are visible in the estimate.
- Phases mirror how you will actually deliver, so scope can be deferred cleanly.
- Roles and seniority mix are realistic for the work, not optimised for price.
- Repeatable elements are priced consistently with your other recent engagements.
- Where a rule or formula produced a number, the rule is stated and the adjustments are documented.
Hidden effort
The lines most often missing from services estimates.
- Project management, coordination, and reporting, as a visible line rather than a percentage buried in rates.
- Internal review and quality assurance of deliverables before they go to the client.
- Client review cycles: the meetings, the consolidation of feedback, the rework.
- Stakeholder management and steering group attendance.
- Environment setup, access provisioning, and waiting for it.
- Onboarding your own team, including any client-mandated training or security clearance.
- Third-party coordination with vendors, incumbents, or other suppliers.
- Travel time, where onsite presence is expected.
Risk and uncertainty
- Every open question is answered, converted into a stated assumption, or excluded.
- No line item is padded in place of an unresolved question.
- The estimate states where confidence is low and what would raise it.
- Contingency, if used, is explicit and attached to named risks rather than spread invisibly.
- You can answer: what would have to be true for this number to be wrong by half?
- Dependencies with an external owner are flagged, with the impact if they slip.
Commercial shape
- The pricing model fits the level of uncertainty; fixed fee on unresolved scope is a margin decision, not a courtesy.
- Blended or role rates match the delivery team you can actually staff.
- Expected margin is calculated on the estimate as it will be delivered, including non-billable coordination.
- Discounts, if any, remove scope or change terms rather than only reducing price.
- Payment milestones align to deliverables the client will accept.
- Licences, infrastructure, and pass-through costs are identified and their ownership stated.
Defensibility
- Each line can be explained in one sentence, in language the client used in discovery.
- Someone who was not in discovery can read the estimate and describe the engagement.
- The narrative document and the estimate describe the same scope, at the same version.
- If the budget were twenty percent smaller, you know what you would defer and the structure allows it.
- You are prepared for the three questions a sceptical buyer asks: why this much, why this long, and what happens if we do less.
After delivery: calibration
- Actual effort by phase and role is compared to the estimate.
- Variances are attributed to a cause: missing scope, wrong driver, client delay, or delivery execution.
- Change orders raised during delivery are reviewed for what discovery should have caught.
- Any driver that repeatedly explained variance becomes a candidate estimating rule.
- Rules that no longer reflect how you deliver are retired, not left in place.
This template is general guidance for scoping and drafting practice, not legal advice. Have your own counsel review contract language before use.
Put this into practice
SOW Studio turns discovery inputs into structured scope, a reviewable estimate, and a client-ready Statement of Work — with your team approving every step.
No credit card required. Full Professional capabilities for 14 days.