Blog
7 min read
An internal proposal approval checklist records who checked a priced quote, what evidence they used and who may authorise its release. It starts after estimating. The question is no longer how to describe the work, but whether this exact offer can leave the agency.
Use the method below as a proposed internal control, not a proven route to higher profit. It is not a contract template. Its rules are choices for your agency, not external requirements that bind agencies generally.
If the brief still contains unanswered questions, start with the questions before a fixed price quote. Use this checklist to decide whether an already-priced offer may leave the agency.
Keep technical approval separate from release authority. A developer can confirm that an integration is feasible without having authority to accept a discount or promise an earlier launch.
Give sales responsibility for assembling the review pack. Include the client file, internal estimate, schedule, supplier evidence and a list of assumptions that affect the offer.
Delivery checks whether the named team can meet the proposed dates alongside existing commitments. A specialist checks technical feasibility. Finance checks how the discount affects contribution within the defined cost boundary. One check cannot replace another.
Choose one named release authority for the quote. This person decides whether the reviewed offer may go to the client within the agency's own authority limits.
A small agency can combine these roles. Record each check separately, even when one director performs several of them. An assumption still needs review when one person holds several roles. Seek another competent opinion if nobody in your agency can test a critical technical claim.
Decide the escalation route before review begins. If the named authority cannot accept the proposed commercial exception, hold the quote until someone with that authority decides.
An urgent client deadline does not give the person preparing the quote permission to bypass a reviewer whose evidence is needed for release. Record the delay instead.
Apply each pass test to the version you intend to release. A tick without evidence is not a completed review. Name a reviewer for every row, including those marked not applicable. Record a reason for each row that does not apply.
| Reviewer | Evidence to inspect | Pass test | Stop reason |
|---|---|---|---|
| Sales lead | Client file and latest negotiation notes | Offer matches the recorded client discussion | An email promises work absent from the file |
| Delivery lead | Dated capacity check and dependency schedule | Named skills and client inputs fit the proposed dates | A required person or input is unavailable |
| Technical specialist | Feasibility note and tests of critical assumptions | Proposed approach works within stated constraints | An untested dependency could change effort |
| Finance reviewer | Estimate, cost rates, supplier quote and discount calculation | Costs reconcile; contribution meets the agency's agreed decision rule | Discount removes contribution or cost evidence has expired |
| Release authority | Exact client file and completed review record | Required checks pass; no release condition remains open | Wrong version, unresolved condition or insufficient authority |
Keep the technical review visible. A commercial director's willingness to accept a lower fee does not answer whether the proposed approach works.
For estimates, the US Government Accountability Office's Cost Estimating and Assessment Guide covers assumptions and risk analysis. It also covers documenting estimates and updating them with actual costs. It was written for government programmes. Use those topics to check the supporting estimate, not as evidence that this agency checklist guarantees accurate quotes.
A useful stop reason names the missing evidence. Write “supplier price expires before our proposed start” rather than “finance concern”. The next reviewer then knows what must change.
Consider an invented agency website quote. All figures exclude VAT. For this example, cost means direct employment cost for planned project hours plus the external testing supplier. It excludes overhead, sales time, financing costs and tax.
Contribution means fee minus that defined cost. It is not net profit. The amount still has to support the costs excluded above.
| Planned work | Hours | Cost per hour | Sell rate per hour | Planned cost | Undiscounted fee |
|---|---|---|---|---|---|
| Design | 40 | £35 | £70 | £1,400 | £2,800 |
| Development | 30 | £60 | £100 | £1,800 | £3,000 |
| External testing | Fixed | Not applicable | Not applicable | £1,000 | £1,000 |
| Total | 70 internal | £4,200 | £6,800 |
The planned contribution is £6,800 minus £4,200, or £2,600. Now sales proposes a £2,600 discount without changing the work. The fee falls to £4,200 and contribution falls to zero, even though the hours and supplier cost remain unchanged.
Stop for review. These invented rates are not benchmarks, and zero is not a recommended acceptance threshold. Your agency chooses its own decision rule. The point is to expose what the discount removes before anyone releases the revised offer.
Finance should show the revised calculation rather than reduce planned hours to preserve the original contribution. Delivery must confirm that any reduction in effort is genuine. A lower price is not evidence of a smaller job.
For a separate retrospective measure, agency project recovery rate compares the service fee with time valued at standard sell rates. Do not substitute that measure for this cost-based release decision.
Use one record per quote, with dated entries for each version. Keep earlier entries rather than overwrite them. The following invented record continues the example above:
Q-041-A.pdf, supported by Q-041-estimate-A. Fee £6,800; defined cost £4,200; planned contribution £2,600.Q-041-B.pdf, supported by Q-041-estimate-B. Requested discount £2,600; revised fee £4,200; planned contribution zero. A remains in the history.If the agreed revision changes again, create version C. Do not approve B in a message and then send an edited file labelled B. Store a fixed copy of the actual reviewed client file beside its supporting estimate.
Before sending, sales checks that the attachment matches the fixed copy named in the release decision, including any accompanying email that changes the offer. Stop if they differ.
“Approved subject to finance” remains held under this proposed method. Silence is not approval. The release authority records a new decision after the condition is closed and the final file has been checked.
Use specific triggers rather than ask whether a revision feels significant. Reopen finance review after any discount, rate or payment-timing change. Ask delivery and the specialist to recheck if the commercial response also changes effort or approach.
Scope changes reopen every review in the matrix. Date changes require delivery to recheck capacity and dependencies. Finance checks supplier price validity and any costs affected by the new dates. A supplier change needs both technical review and fresh evidence of availability and cost.
Every changed offer needs a fresh release decision against its actual file. Reviewers may confirm that their earlier evidence still applies, but they must name the new version. Record why a check needs no further work.
A replacement supplier may charge the same amount but require a different handover missing from your delivery lead's schedule. Price alone cannot clear that change.
Korrel stores cost and sell rates separately within proposal estimates. Keep this release matrix and the final client file in your own review record; the method does not depend on software enforcing it.
For the wider commercial context, the proposal guides cover decisions before and around this release check. Start with your next priced quote, not a rewrite of every proposal.
Create a Korrel account to build proposal estimates alongside your agency's separate release checklist.
RELATED READING:
Use a pre-quote decision tree to test requirements, expose assumptions, and find missing integration facts before you commit to a fixed price.
Design the boundary of each stage before work starts. Define its output, included review, dependencies, and trigger for reopening the fee.
Price discovery as a defined engagement. Cost each output, name the unknowns it will test, and charge only for the risk that remains.
Korrel turns briefs into structured proposals, then tracks what each job actually costs.
START FREE TRIAL