Key Takeaways
- B2B marketers should prioritize credibility signals such as joy sentiment, customer proof, earned media and dark social conversations over reach or volume alone, according to PAN’s Zareen Fidlon
- As AI-generated answers reshape buyer research, brands need citation-ready content, trusted third-party validation and clearer measurement frameworks
B2B marketers are navigating a credibility crisis shaped by artificial intelligence (AI)-driven content saturation, buyer skepticism and dashboards that no longer tell the full story. As generative tools make it easier to produce more content at greater speed, traditional performance metrics such as reach, impressions and click volume are becoming less reliable indicators of trust, influence and revenue impact.
The PAN Brand Experience Report arrives at a critical moment as B2B marketing teams attempt to understand what actually earns buyer conviction. Its findings point to a shift from volume-based marketing to credibility-led engagement, where sentiment quality, real customer proof, dark social conversations and citation-driven visibility increasingly determine whether a brand is trusted, recommended and accurately represented in AI-generated answers.
In this Q&A, Zareen Fidlon, PAN’s EVP, Head of Integrated Marketing and AI Innovation, explains how B2B marketers can adapt to this new trust landscape, discusses how to sequence human, unpolished content with polished conversion assets; and how to reduce brand misrepresentation in AI-generated answers before it shapes buyer perception.
Demand Gen Report (DGR): Zareen, appreciate you taking time to answer our questions. Let’s start here: how should B2B marketers sequence human, unpolished content and polished conversion content in a real campaign so trust builds first and action follows?
Zareen Fidlon: Thanks for having me. Our research found that neither style works on its own. When we ran identical claims to the same CMO audience, unpolished content delivered 10x the click-through rate and held attention 2.3x longer in-feed, but produced zero landing page clicks. Every off-platform action came from the more polished versions. The sequence that works is human content first, like specific anecdotes or plainspoken admissions of friction, to earn dwell time. This should be followed by structured, on-brand content that gives buyers a clear next step. One without the other leaves you spinning in a circle.
DGR: If credibility matters more than volume or polish now, what is one signal B2B marketers should track to tell whether their content is actually building trust instead of just creating noise?
Fidlon: Joy sentiment growth. Across the 20+ B2B brands we analyzed, it was the strongest predictor of revenue, stronger than reach, mention volume or unique author count. It captures something the volume metrics miss: whether the people talking about your brand are saying things that suggest they actually value it. If joy sentiment is moving up while mention volume holds steady or even dips, that’s a brand earning its way into consideration. If reach is climbing and joy is flat or falling, you’re buying visibility without building belief.
Which B2B Trust Signals Matter More Than Reach?
DGR: When reach is up but engagement depth and sentiment are slipping, what are B2B marketers most likely misreading in their dashboards, and which signals should they reinterpret first?
Fidlon: The biggest misread is over-indexing on reach as a success signal. In 40% of the brands we analyzed, reach was growing significantly faster than mention volume, meaning the brand was being referenced more widely but discussed more shallowly. The first signal to reinterpret is the joy-to-negative-emotion ratio. In hypergrowth brands, we saw fear scale roughly three times faster than joy, which often shows up in sentiment before it shows up in revenue. The second is author specificity. A growing unique author count sounds like momentum, but if those authors lack standing in your category, it’s noise. The counter-signal worth internalizing is that some of the brands growing fastest in our dataset were actually declining in social conversation volume.
DGR: If influencer attention does not translate into buyer conviction, especially in enterprise deals, where do influencer programs still fit in a strong credibility strategy?
Fidlon: Our consumer research found a 42-point gap between how buyers value real customer experiences versus influencer content, and influencer endorsement ranked dead last among purchase influences at 14%. The buyers with the most signing authority, Boomers and Gen X, are also the most resistant. That doesn’t mean influencer programs are dead, but the job description changes. They’re attention programs, not credibility programs. They can introduce a brand to a new audience and contribute to top-of-funnel awareness, but the work of building conviction belongs to customer voices and earned editorial. Measure them on whether they create the conditions for a credible second touch, not on engagement alone.
DGR: How can B2B marketers systematically capture and use real customer proof in their marketing without making those stories feel scripted or staged?
Fidlon: Real customer experiences were the top reason participants preferred an ad across two creative preference exercises in our research. The reason most customer stories feel staged is that they’re captured through a process designed to produce something polished. This typically includes interviews funneled into briefs, edited for clarity and approved through legal. The fix is to capture earlier and more often. The most credible customer language usually surfaces in sales calls, support tickets and onboarding conversations rather than scheduled testimonial interviews. Resist the urge to round off the edges; the friction in a story is what makes the rest believable.
How Can Customer Proof Reduce AI Credibility Fatigue?
DGR: What should brands stop doing right now if they want to avoid adding to buyers’ AI credibility fatigue?
Fidlon: Stop producing content whose primary value is volume. If a post could have been generated by any vendor in your category, it contributes to the sameness problem, not stands apart from it. Stop using polish as a substitute for credibility, since the instinct to make every asset look more produced is the same instinct AI-generated content exploits. And stop chasing every new AI tool. The discipline that matters is building practical workflows. 66% of consumers are already experiencing AI credibility fatigue, and 43% say they don’t trust much of anything anymore. The brands making that worse aren’t doing it maliciously, but they’re working from the same playbook that worked five years ago.
DGR: What practical steps should marketing and communications teams take now to reduce brand misrepresentation in AI-generated answers?
Fidlon: Start by auditing what AI is currently saying about you. Run the queries your buyers are running across ChatGPT, Perplexity and Google AI Overviews, and document what’s accurate, what’s wrong and what’s misattributed. From there, prioritize earned editorial in the outlets AI draws from. Earned media accounts for the smallest share of citations by volume at 17%, but it’s the share you can most directly influence and confers the most authority. Structure your owned content to be machine-readable too. It should have clear answers, specific data and named experts with verifiable credentials.
DGR: How should B2B marketing teams listen for and learn from the trust signals forming in dark social channels that standard analytics tools can’t see?
Fidlon: The conviction that moves B2B deals is rarely formed in public. It forms in Slack communities, private LinkedIn groups, Reddit threads and niche forums where buyers recommend vendors and raise objections they’d never voice on a sales call. Start by mapping at least three spaces per persona, then enter as a practitioner rather than a brand. Read for two to four weeks before contributing anything, and track six signals monthly: unprompted brand mentions, competitor comparisons, category frustrations, the trust shortcuts that the community relies on, AI skepticism and what triggers objections to vendor content. Contribute only when you have something genuinely useful to add.
How Can Brands Improve Visibility In AI-Generated Answers?
DGR: How can B2B teams use AI to work faster without letting their message, voice, and point of view blend in with everyone else?
Fidlon: Too many people are using AI at the wrong layer of the work. We’ve built our own internal AI program around this distinction, with one layer that accelerates the high-volume work like drafting, summarization and ideation, and another that surfaces patterns teams can act on strategically. What it deliberately doesn’t do is generate the final output. AI is excellent at compression and expansion, but when teams use it at the output layer, every brand starts to sound like every other brand. The specific anecdote, the contrarian take, the friction your customer admitted to last week, that has to come from a human. If a competitor could publish your content with their logo on it and nobody would notice, you’ve automated yourself into the sameness problem.
DGR: Over the next 12 to 24 months, what shift in buyer behavior or content strategy do you believe most B2B marketers are still underestimating—and why?
Fidlon: Citation infrastructure is replacing traffic as the primary visibility currency, and most teams haven’t reorganized around it. For two decades, the goal of most marketing programs has been driving qualified visitors to owned properties. But 38% of buyers now skip websites entirely because an AI summary already answered their question, and 89% of tech buyers rely on generative AI in purchase decisions. What replaces the destination model is a footprint of citations across earned editorial, analyst reports, structured owned content AI can parse and dark social where peer recommendations form. The reason teams underestimate this is that it doesn’t show up in current dashboards; there’s no clean attribution path from a Forbes mention to a closed deal. The brands that win the next 24 months will be the ones that rebuild their measurement frameworks first.
