Methodology

How we rank influencer marketing agencies

15 agencies are scored on six weighted criteria using public evidence. The calculation is shown on every profile, so anyone can check how a score was reached.

Four numbers we keep separate

Editorial score

A number out of 10, calculated from the weighted criteria below. It is our assessment of public evidence — not a customer rating, and never shown as stars.

Overall rank

An agency's position when all agencies are sorted by editorial score. Ties are broken alphabetically. Demonstration records never receive a rank.

Position in a curated ranking

Pages such as best for TikTok order agencies by fit for one brief. Position there can differ from overall rank; the agency's editorial score does not change.

Third-party review rating

A rating from an external review site, shown only when we have verified it, with the platform, review count, link and date checked. We do not invent or estimate review counts.

Agency scoring criteria

Each criterion is scored 0–10 by the research desk using the evidence types listed. The editorial score is the weighted sum: score = Σ (criterion score × weight) ÷ 100, rounded to one decimal place.

Agency criteria and weights
CriterionWeightWhat countsWhat does not count
Published proof of work

Named clients and case studies with specific, attributable outcomes on a public page.

30%
  • Case studies that name the client, the platform and at least one measured outcome
  • Client names published by the agency or a bylined third party
  • Unnamed "Fortune 500 brands" claims
  • Logo walls with no linked work
Service depth and platform coverage

How much of a creator programme the agency can run in-house, and on which social platforms.

20%
  • Published service pages (strategy, creator sourcing, contracting, production, paid amplification, reporting)
  • Platforms the agency names on its own service pages
  • A list of every social network with no supporting service detail
Footprint and bench

Published offices, markets served, languages and team size band.

15%
  • Office addresses on the contact page
  • Published headcount or a directory headcount band
  • A "global" claim backed by a single office
Ownership and commercial transparency

Whether buyers can see who owns the agency, who runs it, and how to buy from it.

15%
  • Ownership stated on the agency site, an acquisition release or a filing
  • Named leadership, procurement information and published engagement models
  • Ownership we can only infer
Credentials and third-party recognition

Platform partner programmes, industry memberships and awards that can be checked.

10%
  • Partner listings on a platform directory or the agency site
  • Awards with a named programme and year
  • Unlinked badges in a pitch deck
Measurement and reporting

How clearly the agency describes measurement, tooling and accountability for outcomes.

10%
  • Published reporting approach, attribution methods or proprietary tooling
  • Commercial terms linked to agreed KPIs
  • Earned media value presented as the only success metric

Worked example: Socially Powerful

The current leader's score of 9.3 is calculated like this. Every agency profile shows the same table.

Socially Powerful score calculationEach criterion is scored 0–10, multiplied by its weight, and summed. Unknown means no public evidence was found; it contributes 0.
CriterionWeightScorePointsEvidence note
Published proof of work 30% 9.6 2.88 Nineteen published case studies naming client, platform and outcome, including multi-brand L'Oréal USA work.
Service depth and platform coverage 20% 9.5 1.90 Influencer, paid social, strategy, production, community, seeding and commerce published as separate service lines.
Footprint and bench 15% 9.2 1.38 Four staffed hubs across Europe, North America, the Middle East and China; 20+ languages.
Ownership and commercial transparency 15% 9.4 1.41 Independence, founders, CEO and a procurement contact published; headcount figures vary between pages.
Credentials and third-party recognition 10% 8.0 0.80 IMTB membership claimed; awards referenced without itemised programmes and years.
Measurement and reporting 10% 9.4 0.94 ARIA tracking and reporting plus a stated pay-on-KPI guarantee.
Editorial score100%9.3How the calculation works

Labels

  • Leader — the agency with the highest editorial score.
  • Strong performer — editorial score of 8.0 or higher.
  • Specialist pick — below 8.0; often a strong choice for a narrower brief, which curated rankings capture.

How evidence is verified

Verified
Confirmed in an independent or primary third-party source: an acquisition release, regulatory filing, platform directory or named press.
Company claim
Published by the company about itself. We cite it and say so, but have not confirmed it independently.
Unverified / Unknown
Reported but not confirmed, or no public evidence found. When we cannot find public evidence, the field is shown as unknown. Missing public evidence does not prove a capability is absent.

Where an agency's own pages disagree — for example on headcount or founding year — we show the discrepancy rather than choosing the most flattering figure. Client names are recorded as published; a name on a profile does not mean the client endorses the agency or this site.

Platform scoring criteria

Software platforms are scored separately on criteria suited to software and never appear in agency rankings.

Platform criteria and weights
CriterionWeightDescription
Product depth30%Breadth of documented workflow coverage — discovery, relationship management, payments, content rights, reporting.
Customer proof25%Named customers and case studies the vendor publishes.
Pricing clarity15%Whether buyers can see plans, prices and trial terms before talking to sales.
Integrations and ecosystem15%Documented integrations with commerce, social and analytics tools.
Company transparency15%Ownership, location and leadership that a buyer's procurement team can verify.

Independence and disclosures

Positions are not sold. There are no sponsored slots, pay-to-rank fees or affiliate commissions on agency retainers.

Comparisons are only published when we can write a useful case for both agencies. Curated rankings are only published when enough agencies have evidence for the brief.

Common questions

Is the editorial score a customer review rating?

The editorial score is our assessment of public evidence. It is not a customer review rating, and we never convert it into stars.

Can an agency pay to rank higher?

Positions are not sold. There are no sponsored slots, pay-to-rank fees or affiliate commissions on agency retainers.

What happens when evidence is missing?

When we cannot find public evidence, the field is shown as unknown. Missing public evidence does not prove a capability is absent. An unknown criterion contributes zero points, and the profile shows it as unknown so readers can see the gap.

How often are scores updated?

Each record carries its own research, update and next-review dates. The dataset is reviewed at least every 120 days; the next scheduled review is 15 January 2027.

How do I request a correction?

Agencies and readers can request corrections by sending a public URL that supports the change. We log material corrections on the affected page.