01 / HOW WE RANK
Clear criteria.
No mystery scores.
PR Agency Ranking researches agencies against a defined buyer, industry, and brief. An agency must first clear a minimum evidence threshold. Qualified agencies are then assessed across Industry Depth, Campaign Proof, Client Fit, Reputation, and Transparency and Freshness. Scores organize the evidence; they do not replace editorial judgment or guarantee that the highest-ranked agency is right for every buyer.
5 DIMENSIONS
Evidence before scores.
Scores before prose.
Reviewed by: PR Agency Ranking Editorial Desk · Updated: July 16, 2026 · Scoring dimensions: 5 · Commercial policy: No paid inclusion or rank positions
THE FULL FRAMEWORK
Methodology contents
WHAT THE RANKING CAN TELL YOU
What can this methodology tell you?
The methodology is designed to answer a practical question: which agencies have enough relevant, checkable evidence to deserve a place on a buyer’s shortlist for a specific type of PR engagement?
It can show where an agency’s evidence is strongest, which client and brief it appears to fit, what trade-offs are visible, and how confidently the available sources support those conclusions. It cannot predict chemistry, guarantee campaign results, confirm current availability, or replace a direct conversation about scope, budget, conflicts, team access, and measurement.
A ranking position is therefore a research conclusion, not a universal quality label. An agency ranked lower overall can still be the better choice for a particular region, company stage, service need, working model, or budget.
02 / RESEARCH PIPELINE
Our research and publication process
1. Define the buyer and decision
Every ranking starts with a specific search intent, audience, geography, company stage, and use case. A gaming launch, venture fund, early-stage SaaS company, cybersecurity vendor, and consumer technology brand should not receive the same shortlist or scoring emphasis.
2. Build a broad candidate universe
Candidates are discovered through current trade reporting, award records, client announcements, agency portfolios, professional associations, specialist directories, and category research. Appearance in another list is a discovery signal, not evidence of quality and not an automatic qualification.
We aim to research the market before selecting the final list. If a project intentionally evaluates only a bounded sample, agencies outside that sample are labeled as not evaluated rather than rejected.
3. Verify operation and collect evidence
Researchers confirm that each agency appears to be operating and collect dated evidence against all five scoring dimensions. Claims are labeled as verified, supported, self-reported, unverified, or outdated. Unknown information remains unknown; it is not inferred from marketing language.
4. Apply the evidence gate
Agencies that do not meet the minimum evidence threshold are excluded before scoring. This prevents a polished website, a familiar name, or an unsupported client list from competing on equal terms with independently supported work.
5. Set category weights and score
The scoring weights are chosen for the buyer decision before candidate conclusions are reviewed. Researchers then score the usable evidence, apply the evidence-confidence adjustment, document exclusions, and calculate the internal order.
6. Lock the order before writing
The score output is timestamped and locked before narrative drafting. The writing phase cannot move a preferred agency, commercial partner, or recognizable brand into a more favorable position.
7. Write decision boundaries
Every included agency should receive a clear verdict, best-for use case, not-best-for boundary, strengths, trade-offs, evidence explanation, and links to the sources or profile information behind the assessment.
8. Review, publish, and refresh
Before publication, the editorial review checks sources, scoring consistency, conflicts, disclosures, reviewer and update information, links, and unsupported claims. Priority rankings are revisited quarterly and may be reviewed sooner after a closure, acquisition, ownership change, major new evidence, or material correction.
03 / QUALIFICATION GATE
What is the minimum evidence threshold?
An agency qualifies for scoring only when all of the following are satisfied:
- Current operation can be verified.
- A current official agency source is available.
- At least two usable independent sources support the research record.
- At least one relevant client, campaign, or case-study signal exists.
- At least three of the five scoring dimensions contain supported evidence.
- Commercial, ownership, referral, sponsorship, or other relevant conflicts are recorded.
An agency can be credible and still fail the gate for a particular ranking. For example, a respected generalist may lack enough gaming evidence for a gaming ranking, or a specialist may have strong category claims but no independently visible campaign proof.
04 / SOURCE HIERARCHY
Which sources carry the most weight?
| Source level | Examples | How it is used |
|---|---|---|
| Tier A: strongest | Named client confirmation, independent trade reporting, award-body campaign records, company filings, directly observable earned coverage | Can verify important client, campaign, outcome, operating, or reputation claims |
| Tier B: useful | Agency case studies, current client lists, controlled review platforms, professional associations, current leadership and team pages | Supports scoring when specific, current, and corroborated |
| Tier C: discovery only | Generic directories, agency-authored best-of lists, undated roundups, syndicated releases, uncorroborated social posts | Helps find candidates or leads but does not satisfy the independent-source threshold alone |
| Rejected | Scraped listicles, fake review sites, unsourced AI summaries, anonymous claims, unopened search snippets | Not used as evidence |
Self-published evidence is not automatically false. It can establish what an agency claims, describe its services, or provide a case-study lead. It receives less confidence than client confirmation, independent reporting, a reputable award record, or other directly checkable evidence.
05 / SCORING MODEL
What is the default scoring model?
The default model scores five dimensions from 0 to 10 and applies the following weights:
| Dimension | Default weight | Core question |
|---|---|---|
| Industry Depth | 25% | Does the agency demonstrate relevant category experience and understanding? |
| Campaign Proof | 30% | Is there credible evidence of relevant work, execution, and outcomes? |
| Client Fit | 20% | Does the agency’s model fit the buyer’s stage, scope, region, and needs? |
| Reputation | 15% | What do credible third-party signals say about its standing and reliability? |
| Transparency and Freshness | 10% | Is the evidence current, specific, accessible, and responsibly presented? |
Weights total 100%. A category may use a documented variation, but weights cannot be changed after seeing which agency would benefit.
How dimension scores are interpreted
| Score | Meaning |
|---|---|
| 0–2 | Evidence is absent, contradicted, irrelevant, or materially outdated |
| 3–4 | Evidence is weak, narrow, or mainly self-reported |
| 5–6 | A credible baseline exists, but breadth, recency, or independent support is limited |
| 7–8 | Strong, relevant, current, and independently supported evidence |
| 9–10 | Exceptional breadth, depth, recency, and independent proof for the precise category |
High scores require more than name recognition. They require evidence that is relevant to the ranking’s buyer and decision.
06 / FIVE DIMENSIONS
Five ways we test the evidence.
1. Industry Depth
Industry Depth measures whether an agency has meaningful experience in the category being ranked and understands the media, audiences, language, timing, risks, and operating realities that shape the brief.
Strong evidence includes: multiple relevant named clients, current category case studies, specialist team experience, years of sustained work, credible industry participation, appropriate services, and informed perspectives that demonstrate more than surface familiarity.
Weak signals include: generic “we serve this industry” copy, one undated logo, an isolated project with no explanation, category keywords added to a service page, or experience that belongs only to a former employee.
Industry Depth is not a count of logos. A smaller specialist with several well-supported, recent assignments may score better than a large agency with a broad but poorly evidenced client wall.
2. Campaign Proof
Campaign Proof measures whether relevant work can be connected to a named client or campaign, a clear objective, identifiable execution, and a result that can be checked.
Strong evidence includes: client-confirmed work, independent coverage tied to the campaign, award records that describe strategy and execution, named case studies with dates and outcomes, and results corroborated by credible sources.
Weak signals include: logo walls, lists of publications without attribution, screenshots with no campaign context, vague claims such as “generated millions of impressions,” or outcome figures published only by the agency with no supporting detail.
Placements matter, but placements alone do not prove strategic or commercial impact. We distinguish outputs such as coverage volume from outcomes such as qualified interest, reputation change, launch adoption, investor attention, or another stated campaign objective. Campaign Proof cannot score above 5/10 when the entire record is self-reported.
3. Client Fit
Client Fit measures whether the agency’s apparent operating model matches the buyer described by the ranking. It considers company stage, client type, service scope, geography, senior-team access, engagement model, budget accessibility when known, and the practical demands of the brief.
Strong evidence includes: a consistent pattern of comparable clients, clearly described scopes, relevant geographic coverage, a suitable team model, realistic service breadth, and evidence that the agency can support the buyer’s company stage and campaign type.
Weak signals include: universal fit claims, no indication of the clients served, mismatched regions, unclear team access, or a service list that does not show how work is delivered. Unpublished pricing is recorded as unknown, not treated as either expensive or affordable.
Client Fit prevents “best” from becoming synonymous with “largest.” A global network may be wrong for a founder who needs senior boutique access; a specialist boutique may be wrong for a complex multi-market retained program.
4. Reputation
Reputation measures credible third-party signals about an agency’s market standing, leadership, reliability, recognition, and current operation. It also considers material disputes, ownership changes, closures, and other developments that affect buyer confidence.
Strong evidence includes: specific client feedback with review controls, reputable trade recognition, meaningful awards with visible criteria, established leadership records, sustained client relationships, and corroborated market standing.
Weak signals include: raw award counts, pay-to-play badges, anonymous testimonials, review volume with unclear controls, self-described leadership claims, or recognition unrelated to the category being assessed.
Awards are supporting evidence, not a substitute for campaign proof. Reputation also does not become a popularity contest: a newer agency can score well when the available evidence is strong, current, and relevant.
5. Transparency and Freshness
Transparency and Freshness measures whether buyers and researchers can understand who the agency is, what it currently does, and how recent and traceable its evidence is.
Strong evidence includes: current leadership and contact details, dated case studies, accessible client and campaign information, clear ownership and locations, precise claims, recent operational signals, and a responsible corrections process.
Weak signals include: undated work, inaccessible case studies, obsolete team pages, vague superlatives, contradictory locations, unclear ownership, or evidence older than 36 months with no current corroboration.
Older evidence may still provide historical context, but it receives less weight when the ranking is meant to guide a current buying decision. This criterion also makes the methodology more resistant to polished but unverifiable marketing.
07 / EVIDENCE CONFIDENCE
How does evidence confidence affect a score?
The weighted dimension score is adjusted according to the quality of the evidence supporting it:
| Confidence | Multiplier | Typical evidence profile |
|---|---|---|
| High | 1.00 | Verified evidence dominates |
| Medium | 0.90 | Supported evidence dominates, with some gaps or self-reporting |
| Low | 0.75 | Self-reported evidence dominates despite passing the minimum gate |
This adjustment prevents two agencies with superficially similar claims from receiving equal treatment when one record is independently supported and the other depends mostly on agency-controlled material.
08 / CATEGORY WEIGHTS
Why do category weights change?
The five dimensions remain consistent, but different buyer decisions carry different risks. Weight changes are documented in the research brief before conclusions are reviewed.
- Gaming and esports: Industry Depth rises to 35%, Campaign Proof is 25%, Client Fit is 15%, Reputation is 15%, and Transparency and Freshness is 10%.
- Startup and early stage: Client Fit may rise to 30% because accessibility, stage experience, scope flexibility, and senior access materially affect the decision.
- Crisis communications: Reputation and Campaign Proof receive more weight because senior judgment, reliability, and relevant high-stakes experience are central.
- Location-specific rankings: Local market depth is assessed within Industry Depth or through a documented category extension.
The current gaming PR agency ranking shows the gaming variation in practice.
09 / ORDER AND TIES
How are rankings locked and ties broken?
Candidates that fail the evidence gate are removed before sorting. Qualified agencies are ordered by their confidence-adjusted internal score. Ties are broken by Campaign Proof, then Industry Depth, then overall evidence confidence.
The order is locked before prose drafting. Fit labels such as “best for indie launches” or “best for global publisher programs” must follow the evidence and cannot be created by changing the score.
We do not publish false precision. Internal decimal scores may remain private when source differences cannot support a meaningful public decimal. If the evidence supports a useful shortlist but not a defensible ordinal order, we should publish a clearly labeled non-ordinal shortlist instead of forcing numbers onto the page.
10 / WORKED EXAMPLE
Worked scoring example
The following example is hypothetical and does not describe a real agency.
| Dimension | Score | Default weight | Weighted contribution |
|---|---|---|---|
| Industry Depth | 8.0 | 25% | 2.00 |
| Campaign Proof | 7.0 | 30% | 2.10 |
| Client Fit | 6.0 | 20% | 1.20 |
| Reputation | 7.0 | 15% | 1.05 |
| Transparency and Freshness | 8.0 | 10% | 0.80 |
| Weighted score | 100% | 7.15/10 |
If supported rather than verified evidence dominates, the example receives the medium-confidence multiplier: 7.15 × 0.90 = an internal confidence-adjusted score of 6.44/10. The published review would still explain the underlying fit, evidence, strengths, and limitations rather than presenting the number as a prediction of results.
11 / PUBLISHED DECISION BOUNDARIES
What does every published agency review include?
Each substantive agency review should answer:
- What is the agency best suited for?
- Who should consider another option?
- Which clients, campaigns, services, locations, or team signals are supported?
- Where is the evidence strongest?
- Which claims are self-reported, unknown, limited, or old?
- What practical trade-offs should a buyer investigate?
- When was the information last reviewed?
This structure helps buyers challenge the ranking instead of treating it as a universal recommendation.
12 / COMMERCIAL SAFEGUARDS
Can agencies pay for inclusion or ranking position?
No. Agencies cannot purchase organic inclusion or rank position. Submitting information does not guarantee evaluation, inclusion, or placement. Paid visibility, if introduced, must remain visually and editorially separate from organic rankings.
PR Agency Ranking may have commercial, client, affiliate, referral, sponsorship, or other relationships with an evaluated agency. Those relationships must be recorded before scoring and disclosed on every affected ranking. The same evidence gate, weights, scoring rules, and correction access apply to conflicted and non-conflicted candidates.
Read the commercial disclosures, editorial policy, and corrections policy for the complete safeguards.
13 / USING AND MAINTAINING THE RANKINGS
How should buyers use a ranking?
Use a ranking to create and challenge a shortlist, not to outsource the final decision. Confirm the agency’s availability, budget fit, conflicts, proposed team, senior access, measurement plan, contract terms, and working style directly.
Ask each shortlisted agency to explain who will do the work, which comparable campaigns it can substantiate, how success will be measured, what is outside scope, and which references can be checked. A strong proposal should make its assumptions and trade-offs visible.
How often are rankings reviewed?
Priority rankings are reviewed quarterly. A page may be checked sooner when an agency closes, merges, changes ownership, loses a material capability, publishes substantial new evidence, or receives a correction that changes buyer understanding.
Stable URLs are retained during refreshes. Update dates are changed only after substantive review. Revision history remains in private editorial records; a public correction note is added when a material change affects the reader’s understanding of the ranking.
14 / QUESTIONS BUYERS ASK
Frequently asked questions
Does the highest score mean the agency is best for everyone?
No. The score reflects the defined category and buyer. A lower-ranked agency may fit a specific budget, region, company stage, specialist need, or engagement model better.
Why are some well-known agencies missing?
An agency may fall outside the research scope, fail the category-specific evidence gate, lack enough current sources, or be less relevant to the precise buyer being assessed. Absence is not automatically a judgment about the agency’s overall quality.
Do agency case studies count as evidence?
Yes, but they are agency-controlled sources. They carry more weight when they name the client, objective, execution, date, and outcome and when important claims can be independently corroborated.
Why are public scores sometimes withheld?
Evidence can support a defensible order without supporting meaningful decimal precision. In that situation, we publish the rationale, fit, and trade-offs while retaining detailed calculations in the private editorial record.
Can an agency correct its profile or ranking entry?
Yes. Agencies and readers can submit traceable evidence through the corrections process. A correction does not guarantee a ranking change; the new evidence is assessed under the same rules.
Who reviews the rankings?
Published work identifies its reviewer and update date. Until a named human reviewer is publicly confirmed, methodology and foundation pages are reviewed by the PR Agency Ranking Editorial Desk.
15 / AUDIT THE SYSTEM
Sources and editorial basis
This page documents PR Agency Ranking’s own research and scoring protocol. Its operational references are the editorial policy, commercial disclosures, corrections policy, current rankings hub, and the published gaming ranking where the category-specific weighting is applied.