Reviews
Part of How to run influencer agencies reviews fairly
Influencer agencies review methodology: gates and scoring
Use this influencer agencies review methodology to assess UK agency identity, campaign scope, advertising, rights, data handling and buyer-set scoring.
Influencer agencies review methodology: gates and scoring
In this influencer agencies review methodology, first decide whether one precisely scoped service is admissible before assigning preferences. For a UK commission, record the contracting entity, service version, campaign task, evidence cut-off and required gates. Define the work being bought—such as creator sourcing, campaign strategy and management, content or user-generated content (UGC) production, paid-social amplification or whitelisting, platform-specific execution, usage rights and reporting—because an agency's remit varies by brief. A polished pitch cannot compensate for missing authority, privacy or rights evidence.
Reviewer disclosure: creatorcampaign.co.uk designed this method. It received no supplier funding and contains no affiliate relationship. It was desk research only: no agency contact, reference call, platform login, creator discussion or live test occurred.
What to take away
- A review method must decide if a service is admissible before assigning any preferences.
- Mandatory gates use PASS, HOLD or FAIL and are never averaged together.
- Blank weights for admitted candidates are set before opening price proposals.
- Evidence must be listed with version and date, and supplier claims stay attributed until verified.
Define the subject and evidence set
Start a dossier with legal name, trading name, company number where applicable, contracting jurisdiction, named service, availability by UK nation, channels, deliverables, exclusions, data access and quote date. Companies House's GOV.UK company-information service supports checking registered details and filings. Treat it as identity evidence only, not proof of capability or solvency.
List every document opened, URL, version and date. Separate official records, signed buyer-supplier material, permissioned references, controlled buyer tests, first-party descriptions, case studies and testimonials. Supplier examples remain attributed claims until the buyer verifies the underlying scope and permission.
Candidate lists for an England buyer come from trade bodies such as the IPA, ISBA and PRCA, plus specialist influencer directories and this site's own agency directory.
Examples of agencies a UK buyer could research include Billion Dollar Boy, Goat, Influencer and Socially Powerful. Their inclusion is not a review or endorsement, and does not imply that they offer the same services; verify each agency's current entity, campaign scope and availability through the method below.
Apply a common matrix
For each candidate and the in-house manual baseline, use PASS, HOLD or FAIL on these mandatory rows:
Mandatory Agency Review Gates
- Identity and authority
- Creator provenance
- Advertising and claims
- Rights
- Safeguarding
- Privacy and PECR
- Accessibility
- Security and ownership
| Gate | Evidence request | Stop condition |
|---|---|---|
| Identity and authority | Entity record, signatory power, creator or talent mandate | Contracting party or authority is unresolved |
| Creator provenance | Source, permission, field definitions, currency and correction route | Origin or permitted use cannot be shown |
| Advertising and claims | Brief, substantiation, disclosure, approval and withdrawal workflow | Objective claim or ad recognition has no owner |
| Rights | Asset, music, performance, edit, territory, term and media schedule | Intended use exceeds evidenced permission |
| Safeguarding | Age checks, escalation and England licensing review | A relevant child-work question is unanswered |
| Privacy and PECR | Purpose, role, outreach, tracking, retention and rights map | Processing or communication route lacks review |
| Accessibility | Format criteria and final-render evidence | Required user journey was not checked |
| Security and ownership | Buyer accounts, access log, recovery and revocation test | Shared or non-recoverable control remains |
| Commercial | Payer, consideration, tax evidence, fee unit and conflicts | Money or benefit flow is opaque |
| Measurement and exit | Metric cards, exports, deletion, continuity and corrections | Evidence cannot be reconstructed or removed |
The CMA's Social media endorsements: being transparent with your followers guidance supports a distinct briefing, checking and correction process; it does not give an agency a quality mark. Advertising is governed by the UK Code of Non-broadcast Advertising and Direct & Promotional Marketing (the CAP Code), administered by the Committee of Advertising Practice and enforced by the Advertising Standards Authority (ASA); a complaint to the ASA may result in an ASA ruling. For personal data, assess the UK GDPR and Data Protection Act 2018: the ICO's controller and processor guidance requires roles to follow actual decisions, so a supplier's contract label is not enough. A person can complain to the ICO about a data-protection concern.
Score only admitted candidates
Do not average mandatory gates. If one is HOLD or FAIL, record the reason and stop the weighted comparison.
For passing candidates, the buyer may assign blank weights to evidence quality ___, operating fit ___, total economic cost ___, support ___ and exit resilience ___. For cost, compare quotes for the same scope and record actual quoted retainers, campaign or management fees, commission, creator fees, production, usage or whitelisting, paid media and exit charges where applicable; do not treat an omitted item as zero. Define scoring anchors before opening price proposals, and have two reviewers resolve disagreements against stored evidence.
Example anchors: 1 = no evidence, 3 = partial evidence with gaps, 5 = complete, dated and permissioned evidence. The following sample weight split is illustrative only, not fixed or prescribed: evidence quality 30, operating fit 25, economic cost 20, support 15, exit resilience 10. Buyers should replace the blank fields with their own weights before opening proposals.
In a hypothetical illustration, not an assessment of real suppliers, Agency A fails the rights gate because its media schedule omits paid social rights; it stops at FAIL. Agency B passes every gate. The buyer scores B as evidence quality 4, operating fit 5, economic cost 3, support 4, exit resilience 2. Its weighted total is 3.85 out of 5.
A test result must name the tester, date, permissioned environment, script, synthetic or real data, expected outcome, observed outcome and limitation. NCSC social-media account guidance supports checking controlled access and audit trails. A demonstration watched by the buyer is not a hands-on recovery test.
State the verdict narrowly
PASS means the exact dossier met the buyer's stated gate on the review date. It does not mean legally compliant, best in England or likely to deliver results. HOLD means evidence is incomplete; FAIL means a predefined requirement was not met. Set an expiry and recheck trigger for ownership, service, platform, regulatory or contract change.
For a live UK appointment, include procurement, advertising, privacy and PECR reviewers, as well as IP, safeguarding, accessibility, security, tax, contract and evaluation reviewers. For child performers, identify the relevant local authority's licensing requirements for the UK nation where work takes place; in England, check the applicable Children (Performances and Activities) (England) Regulations with the local authority. This framework is not professional advice.
Before you act
- Create a blank dossier row for each candidate with contracting entity, service version and evidence cut-off.
- Record PASS, HOLD or FAIL for each mandatory gate in the matrix above.
- Stop the weighted comparison as soon as one gate is HOLD or FAIL.
- Before opening any price proposal, write down the five weights and the 1 to 5 scoring anchors.
- Set a recheck date and a trigger for ownership, platform or regulatory change.
Common questions
What must be recorded for an England commission before preferences are assigned?
For the England scenario, record the contracting entity, service version, campaign task, evidence cut-off and required gates. For a UK-wide commission, also record each nation where work occurs and check nation-specific requirements. A polished pitch cannot compensate for missing authority, privacy or rights evidence.
How should mandatory gates be handled in the common matrix?
Use PASS, HOLD or FAIL on each mandatory row. Do not average mandatory gates. If one is HOLD or FAIL, record the reason and stop the weighted comparison. Only candidates that pass may be scored with blank weights.
