Strategy
Part of Before you shortlist branding agencies: decide whether to commission at all
A decision record for assessing branding agency claims
Use this branding agencies strategy framework to assess a profile’s claim: record its evidence, identify unknowns and decide what it does—and does not—support.
When a branding agency profile or review makes a strategy claim, assess it against a decision record. This strategy framework turns a positioning phrase into questions about evidence, options, measures and review gates.
This framework is for an England-facing commercial decision about a branding agency strategy claim, whether encountered in a profile, review or proposal. It names no agency and quotes no fee: those must come from the evidence gathered for each option. It borrows option discipline from the Green Book, the government's guidance on assessing the costs, benefits and risks of different options. That guidance is written for government appraisal, so private agencies should treat it as a method, not a mandate.
Use agency profiles and reviews as evidence
A directory profile or agency review can help identify a strategy claim, but it is not the decision record itself. Record the claim's exact wording, promised outcome, audience, scope and timeframe. Then note the evidence's source and date, how directly it measures that outcome, whether it covers the same audience and service, what baseline or comparison exists, and what limitations or alternative explanations remain. Distinguish the agency's own claims from a reviewer's account; state what the evidence does—and does not—support.
What to take away
- Use a decision record to assess a branding agency's strategy claim: state the outcome, whose situation matters, which choices remain open and what evidence would make the owner stop.
- Name the business decision the claim would inform, one accountable owner and a decision date.
- Compare complete options, including continuing as now and improving the manual route.
- Keep legal and safety reviews separate from commercial scores; they can veto any option.
- If evidence is weak, continue the current route and commission missing research.
Write the decision before the ambition
Name the buyer's decision about the agency's strategy claim, one accountable owner and the date on which that person will decide. Then complete these fields:
Decision record fields
- Objectiveobservable change, eligible audience
- Baselinecurrent process, dated evidence
- Unknownsfacts records cannot answer
- Constraintscash, capacity, rights, timing
- Optionscontinue, improve, commission, build
- Acceptance measurenumerator, denominator, period
- Stop ruleresult, harm, missing approval
- objectivethe observable change sought, for a defined eligible audience;
- baselinethe current process and dated evidence showing its condition;
- unknownsfacts that the existing records cannot answer;
- constraintscash, capacity, rights, timing and non-negotiable review gates;
- optionscontinue as now, improve the manual route, commission a partner, or build a new capability;
- acceptance measurenumerator, denominator, population, period and evidence owner;
- stop rulethe result, harm, missing approval or data defect that ends the option.
Worked example, illustrative—not a real agency profile. Imagine a profile claims: “Its strategy will help food and drink SMEs secure five packaging projects.” The record below is an agency-side service plan, not evidence of that client outcome.
Objective: five packaging projects from food and drink SMEs. Baseline: 14 relevant enquiries in six months. Unknowns: willingness to pay a premium. Constraints: two designers from July. Options: continue, manual improvement, freelance partner, recruit. Measure: signed proposals divided by qualified enquiries. Stop rule: fewer than three signed proposals by 31 December.
This record can help assess whether the agency's proposed service appears commercially viable to it. It does not substantiate the profile's claim about client outcomes: an internal count of enquiries and signed proposals is not evidence that clients secured packaging projects through the agency's strategy, and willingness to pay remains unknown. Ask for client-side evidence matched to the claim's audience, service, baseline, measure and timeframe; otherwise, treat the outcome claim as unsubstantiated. This is an illustrative assessment, not evidence about any agency.
Do not turn a persona into a fact. If the accepted research design has not produced evidence, label the audience description a hypothesis and state how it will be tested.
In government service design, teams keep researching users throughout each development phase. Borrow that habit as method: it does not prove how private-sector buyers behave.
Establish a bounded baseline
England is not a convenient synonym for the UK. The latest ONS release counted 2.79 million VAT and/or PAYE businesses in the UK in March 2026, up 1.9% on March 2025, and found professional, scientific and technical activity to be the largest industry group at 15.3% of registered businesses.
These are UK-wide counts by broad industry. They do not isolate branding buyers, agency demand or an addressable market, so use them only as a dated frame for sampling, never as a demand figure.
Baseline record fields
- Source of the baseline record
- Release date of the source
- Geography covered
- Unit of measurement
- Exclusions and known error
- Leave blank if not lawfully retrievable
Record the baseline source, release date, geography, unit, exclusions and known error. Operational evidence might instead be a versioned enquiry log, proposal register or delivery ledger. Leave the field blank if the owner cannot lawfully retrieve a reliable record.
Compare complete options
Apply the same questions to every route under consideration in response to the agency's strategy claim. Can it meet the objective? What has to be true? Which resources and dependencies does it need? For each piece of supporting evidence, ask whether it directly measures the claimed outcome and matches the audience, service and timeframe.
Who carries delivery and exit risk? What evidence would reject it? Keep the current arrangement and a manual improvement as real comparators, and record fees for each route using the pricing models as a guide. A preferred agency engagement should not win merely because alternatives were never described.
The Magenta Book is HM Treasury guidance on what to consider when designing an evaluation. Used as method, not mandate, it points to specifying what will be observed before commissioning work, and to avoiding a causal claim unless the design can support one.
Keep an assumption register beside the option cards. Each entry should state who made the assumption, what observation could disprove it, who will obtain that evidence and when the decision must be reopened. An assumption without a feasible check is a constraint or an unknown, not a confident planning input.
Put reviews outside the score
Commercial attractiveness cannot offset a failed legal gate. Assign a named reviewer to each gate in the table, and check advertising evidence before relying on a claim in an agency profile, review or pitch.
| Gate | Rule or authority | What the reviewer confirms |
|---|---|---|
| Trade marks | Intellectual Property Office | The name is not the same as or similar to an existing mark. Registration costs at least £205 and lasts 10 years |
| Advertising | CAP non-broadcast code, enforced by the Advertising Standards Authority | Online, print, sales promotion and email claims follow the code |
| Privacy and direct marketing | UK GDPR and PECR, regulated by the ICO | Marketing emails or texts to individuals have specific consent. The data protection fee is paid unless exempt |
| Accessibility and equality | Equality Act 2010 | The Act protects people from discrimination in the workplace and wider society, so hiring and service terms follow it |
| Security | Cyber Essentials | Standard technical controls are in place against common online threats |
| Tax | VAT rules | VAT registration is due once taxable turnover passes £90,000 in 12 months |
Record whether the strategy claim or proposed route is approved, held or rejected, with the evidence inspected, using the selection checklist as the template. No score from the strategy workshop guarantees compliance.
The decision owner should review the record when a core assumption fails, a source changes, the eligible audience changes or a reviewer places a hold. If evidence remains too weak, continue the current or manual route and commission the missing research. That is a decision, not indecision.
Before you act
- Name the business decision, one owner and a decision date.
- Record the agency's exact published claim separately from its supporting evidence; label audience descriptions as hypotheses if evidence is missing.
- Check whether evidence matches the claim's outcome, audience, service and timeframe; record its source, baseline and limitations.
- Record baseline source, release date, geography, unit and exclusions.
- Compare all options using the same questions.
- Assign named reviewers for legal, privacy, accessibility and finance gates.
- Review the record when a core assumption fails or a source changes.
Common questions
What fields should a decision record contain?
It should contain the exact claim being assessed, objective, baseline, unknowns, constraints, options, acceptance measure and stop rule. The objective is the observable change for a defined eligible audience. The baseline is the current process and dated evidence. Check whether supporting evidence measures the same outcome for the same audience and service, and record its limitations. Options include continue as now, improve the manual route, commission a partner or build a new capability.
When should the decision owner review the record?
The decision owner should review the record when a core assumption fails, a source changes, the eligible audience changes or a reviewer places a hold. If evidence remains too weak, continue the current or manual route and commission the missing research. That is a decision, not indecision.
