Executive Summary

• Guaranteed earned media conflicts with editorial independence and often masks paid or low-credibility placements, which carry brand and ROI risks.[1][2][3]
• AI-driven discovery strengthens the case for genuine earned media: more than 95% of AI-cited links are non-paid; about 89% are earned; journalism comprises 27% overall and rises to 49% in recency-oriented queries (Muck Rack, 2025).[4]
• Success-fee models tied to outputs (coverage counts, EMV/AEV) create misaligned incentives and measurement distortion; replace with controllable inputs and validated business outcomes.[5][6][7]
• Practical fee structures: base + process SLAs; base + qualified opportunity bonuses; base + analytics-validated outcome bonuses; dual-track earned and transparently labeled paid.[3][8][5]
• Align to AI dynamics: prioritize high-authority, niche-relevant outlets; build a “recency engine” (credible data/news hooks); separate paid from earned to preserve trust and AI visibility.[4]

• Why “Guaranteed Coverage” Doesn’t Square With Earned Media Reality

• Earned media rests on independent editorial decisions, so agencies cannot ethically guarantee coverage or framing.[1]
• “Guaranteed media coverage” usually means paid/sponsored placements or low-authority sites, which erode credibility and yield weak business impact.[2]
• Industry debate acknowledges demand for outcome-based fees but underscores the uncontrollable nature of editorial judgment and timing, making guarantees misaligned with the medium.[3]
• AI context: Generative systems overwhelmingly cite non-paid sources—more than 95% of links—and emphasize earned media (~89%), with journalism comprising 27% overall and 49% for recency-seeking queries, reinforcing the strategic value of genuine earned coverage over pay-to-play (Muck Rack, 2025).[4]

• Core Critiques of Success-Fee/Outcome-Based PR

• Misaligned incentives and adverse selection: Paying for coverage volume/outlet tiers pushes toward pay-to-play, contributed ads, or lower-quality outlets, risking brand trust and long-term value.[2]
• Measurement distortion: Outputs like EMV/AEV are subjective and not reliably linked to business outcomes, incentivizing quantity over impact.[5]
• Ethical and transparency risks: Undisclosed sponsored content blurs paid vs. earned distinctions, undermining editorial integrity.[2]
• Uncontrollable dependency: Editorial calendars and reporter discretion sit outside agency control; contingent fees shift uncontrollable risk to agencies and encourage gaming.[3]
• AI-era implication: Paid/advertorial content is far less likely to be cited by AI compared to journalism, diminishing long-tail discoverability from pay-for-coverage schemes (Muck Rack, 2025).[4]

• Evidence and Analogs From Pay-for-Performance Literature

• Healthcare P4P shows consistent pitfalls when incentives target narrow outputs: gaming, uncertain quality improvement, and ethical concerns—cautionary for PR models tied to coverage counts or EMV.[6][7]

• Squaring Client Demand With Reality in 2025

• Guarantee the process, not the placement:
• Commit to controllables: audience/beat mapping, narrative and data asset creation, pre-briefs, embargo strategies, media training, and rigorous follow-through—“guaranteed process” to maximize odds without promising editorial outcomes.[8]
• Align to AI dynamics: prioritize high-domain-authority and niche-appropriate outlets (e.g., Reuters, AP, FT, Axios are frequently cited), improving both newsroom acceptance and AI citation odds (Muck Rack, 2025).[4]
• Replace EMV/AEV with business-linked outcomes:
• Measure article-level referral traffic, assisted conversions, demo lift post-coverage, share-of-voice vs. a defined set, message pull-through, and qualified awareness/sentiment shifts—avoid EMV/AEV.[5]
• Build recency into the editorial calendar (e.g., quarterly proprietary data studies) because AI prioritizes fresh, topical journalism, especially for event-driven queries (Muck Rack, 2025).[4]
• Separate earned and paid with explicit transparency:
• Use labeled paid/native for speed or scale; maintain distinct KPIs and disclosures.[2]
• Recognize paid has limited AI citation value relative to earned journalism (Muck Rack, 2025).[4]
• Structure fees around controllables and validated outcomes:
• Base retainer for strategy, asset creation, and sustained media relations.[3]
• Milestone bonuses for asset delivery, pre-briefs completed, interviews scheduled/accepted by editorial desks—measured as “booked,” not “published”.[8][3]
• Outcome bonuses tied to analytics-based KPIs within agreed attribution windows (article-driven sessions, assisted conversions, brand search lift, message pull-through), explicitly excluding EMV/AEV.[5]
• Codify risk-sharing guardrails:
• Define required client inputs (spokespeople availability, embargoed news, customer references, proprietary data); renegotiate if inputs lapse.[3]
• Tailor outreach to industry patterns observed in AI: government/NGO prominence in healthcare; heavy journalism weighting in finance, media/entertainment, and technology; niche-relevant sources matter across sectors (Muck Rack, 2025).[4]

• Practical Fee Model Options

• Base + Process SLA:
• Fixed monthly fee plus SLAs on controllable activities: targeted pitches, pre-briefs, rapid response participation, contributed bylines, proprietary data assets—auditable in reporting.[8][3]
• Add a “recency engine”: recurring, credible news hooks/data drops to match AI’s recency bias (56% of journalism citations within 12 months for OpenAI models vs. 36% for Anthropic in the sample) (Muck Rack, 2025).[4]
• Base + Qualified Opportunity Bonuses:
• Bonuses for journalist-accepted interviews/briefings or editorial acceptance of contributed articles by named outlets—rewarding progress points with higher agency control.[3]
• Weight bonuses toward high-authority and niche-relevant outlets favored in AI citations (Muck Rack, 2025).[4]
• Base + Analytics-Validated Outcome Bonuses:
• Bonuses triggered by agreed business outcomes in defined post-publication windows: article-driven traffic, assisted conversions, branded search lift, message pull-through scores; exclude EMV/AEV.[5]
• Secondary objective: improve AI discoverability by targeting reputable, industry-appropriate sources most cited in the client’s vertical (Muck Rack, 2025).[4]
• Dual-Track Earned and Paid:
• Run earned alongside a clearly labeled paid program for time-sensitive goals; distinct KPIs acknowledge that earned fuels credibility and AI visibility, while paid delivers predictable reach (Muck Rack, 2025).[2][4]

• How to Push Back—Concisely and Credibly

• No ethical guarantees: Editors and journalists decide what runs; “guaranteed coverage” typically means paid or low-credibility placements that don’t deliver durable value.[1][2][3]
• Fix the metrics: Outputs like EMV/AEV incentivize quantity over impact and are widely criticized; link incentives to validated, analytics-based outcomes and high-quality controllables.[5]
• Optimize for the AI era: Over 95% of AI-cited links are non-paid and ~89% are earned; journalism is 27% overall and 49% for recency queries—invest in credible, timely earned coverage to strengthen trust and AI visibility (Muck Rack, 2025).[4]
• Offer a guaranteed process and hybrid incentives: Tie bonuses to accepted interviews, analyst mentions, and post-coverage business signals, aligning expectations with editorial independence and AI-driven discovery.[8][3][4]

Appendix: Implementation Checklist

• Editorial engine:
• Quarterly proprietary data report with original stats and visuals.
• Spokesperson media training and availability SLAs.
• Rapid response protocol mapped to target beats and timing windows.
• Embargoed pre-brief calendar for major releases.
• Measurement and attribution:
• UTM strategy for article-level analytics and assisted conversion tracking.
• Message pull-through scoring rubric and third-party coding plan.
• Share-of-voice vs. named peer set in target markets.
• AI-era targeting:
• Priority lists of high-authority and niche-relevant outlets per vertical.
• Recency cadence and freshness SLAs to align with AI citation patterns.
• Clear separation of earned vs. paid placements in reporting.

Full Sources

• Muck Rack. Generative Pulse 2025: What Is AI Reading? (Attached PDF: MuckRack-GenerativePulse2025.pdf). Key findings cited: “More than 95% of links cited by AI are non-paid coverage,” “More than 89% of links cited by AI are earned media,” “27% of links cited by AI are journalistic,” “49% of links cited by AI are journalism for recency-oriented queries,” “Outlet authority matters,” and “AI systems prefer stories in the last 12 months (56% OpenAI vs. 36% Anthropic within sample).”[4]
• Harvard Business School Online. “Paid vs. Owned vs. Earned Media: What’s the Difference?” Overview of earned media as independent editorial exposure.[1]
• Brito, Michael. “Earned Media Value: Don’t Do It!” Critique of EMV/AEV and alternatives for impact measurement.[5]
• Axia PR. “Is ‘guaranteed media coverage’ from PR firms too good to be true?” Analysis of guarantees often translating to paid/lookalike placements and risks.[2]
• PRWeek. “Should PR agencies guarantee earned media? Industry debates.” Trade debate on guarantees and structural limits of earned media.[3]
• Society of General Internal Medicine (SGIM). “High Quality Care and Ethical Pay-for-Performance: A Society of General Internal Medicine Position Statement.” Cautionary insights on P4P metrics and gaming risks.[6]
• RTI Press. “Pay for Performance in Health Care: Methods and Approaches.” Overview of P4P methods and pitfalls useful as analogs for PR incentives[7]

1. https://online.hbs.edu/blog/post/earned-vs-paid-media    
2. https://www.axiapr.com/blog/is-guaranteed-media-coverage-from-pr-firms-too-good-to-be-true        
3. https://www.prweek.co.uk/article/1887024/pr-agencies-guarantee-earned-media-industry-debates            
4. MuckRack-GenerativePulse2025.pdf               
5. https://www.britopian.com/social-data-analytics/earned-media-value/        
6. https://pmc.ncbi.nlm.nih.gov/articles/PMC2695523/   
7. https://www.rti.org/rti-press-publication/pay-performance-health-care-methods-approaches/fulltext.pdf   
8. https://leapshq.com/blog/guaranteed-media-placements