14 min read · March 2026
PR in the Age of AI Search
What every communications professional needs to know about the data reshaping search, citations, and visibility
By Lexi Mills — Built on research by people she deeply respects, starting with Ross Simmonds & the team at Foundation
The Bottom Line
Four stats from a single study that should change how every PR professional thinks about search, visibility, and earned media in 2026.
Monthly searches where a third-party source—press coverage, Reddit, independent reviews—reaches buyers before any brand’s own content.1
Third-party sources outperform brand-owned content across three of four B2B verticals studied. Earned media isn’t just nice to have—it’s the dominant discovery channel.2
Of the search volume that third-party sources win, 77% comes from non-review keywords. This isn’t about “best of” lists—it’s about fundamental category terms.3
At 6+ words—the conversational queries people ask AI search and LLMs—third-party content wins 73–100% of the time across verticals.4
All statistics from Foundation Inc’s analysis of 8,566 keywords across 13 B2B SaaS domains by Ross Simmonds (March 2026)
Why This Matters for PR
Foundation’s report is written for marketers and SEOs. But its implications for PR and communications professionals are arguably even bigger. Here’s the translation.
The Core Insight: Third-Party Content Wins
When Ross Simmonds says “Reddit is eating B2B search,” what PR people should hear is: third-party, independent, earned content is beating brand-owned content at massive scale.
Reddit threads are, in essence, unpaid testimonials, community discussions, and independent recommendations. In PR terms, that’s earned media. And the data says earned media outperforms owned media by 50–67% on the keywords that matter most.
“Third-party sources” in this data = press coverage, earned media, independent reviews, community discussion. “Own content” = brand website, corporate blog, product pages. The data says your press coverage, analyst reports, and media placements carry more weight in search than anything your content team publishes on your own domain.
PR Is Best Positioned to Influence LLM Citations
Here’s what most people miss: AI search systems (ChatGPT, Perplexity, Google AI Overviews) don’t just pull from brand websites. They synthesize information from the sources they trust most—and those sources are overwhelmingly third-party publishers, news outlets, and community discussions.
This is exactly the territory PR professionals operate in. Media placements, analyst briefings, thought leadership, community engagement—these are the inputs that shape how LLMs understand and represent your brand.
If 77% of winning search volume comes from generic category terms—not “best X” or “X review”—then the content shaping brand perception is foundational, not evaluative. It’s the press coverage explaining what a category is, the analyst report defining market boundaries, the community thread where someone asks “how does this type of software work?” Those are PR conversations, not marketing campaigns.
Long-Tail = Conversational = LLM Queries
Foundation’s data shows that as search queries get longer, third-party content dominance increases. At 6+ words, the win rate hits 73–100%. This matters enormously because AI-powered search is pushing all queries in that direction: longer, more conversational, more specific.
When someone asks an LLM “what’s the best CRM for a small business with invoicing and project management,” the answer won’t come from Salesforce.com. It’ll come from whoever created the most trusted, specific third-party content about that exact question. That’s a PR play, not an SEO play.
What the Data Shows
Foundation analysed 8,566 keywords across 13 major B2B SaaS domains in four verticals: Review Sites, SaaS Platforms, Sales Tech, and UCaaS/CCaaS. Here are the numbers that matter most for PR professionals.
Reddit’s Dominance by Vertical
In three of four verticals, third-party content (represented by Reddit) commands 40–45% of top-3 search positions—leading every vendor.
| Vertical | Reddit Top-3 Share | Best Vendor Top-3 | “Beats All” Rate | Threat Index |
|---|---|---|---|---|
| Sales Tech | 41.1% | ZoomInfo: 10.2% | 66.5% | 93/100 |
| Review Sites | 45.4% | Trustpilot: 29.7% | 50.3% | 72/100 |
| SaaS Platforms | 39.8% | Salesforce: 25.3% | 53.3% | 58/100 |
| UCaaS / CCaaS | 10.5% | RingCentral: 23.2% | 17.3% | 22/100 |
Source: Foundation Inc, “Beats All” = Reddit outranks every vendor simultaneously on a keyword. Threat Index composites Beats-All rate, traffic share, keyword value share, and difficulty crossover.5
It’s Not Just Reviews
The most common assumption is that Reddit wins on “best X” and “X alternatives” queries. The data says otherwise.
| Keyword Type | Keywords | % of Dataset | Reddit Win Rate | Share of Won Volume |
|---|---|---|---|---|
| Evaluation (review, best, vs, compare) | 2,172 | 25.4% | 63.4% | 23%* |
| Generic Category Terms | 6,394 | 74.6% | 44.5% | 77% |
*23% / 77% reflects share of Reddit’s won search volume; evaluation keywords account for 32.6% of Reddit’s wins by keyword count.3
77% of the search volume that third-party sources win comes from generic category keywords—terms like “email marketing software,” “CRM for small business,” and “sales automation tools.” These are the foundational, category-defining queries. The conversations that shape how an entire market segment understands what solutions exist. This is classic thought leadership and media relations territory.
The Long-Tail Escalator
Every additional word in a search query increases the probability that third-party content wins. At 6+ words—the type of conversational queries people type into AI search tools and LLMs—the advantage becomes near-total.
| Query Length | Sales Tech | SaaS Platforms | Review Sites | UCaaS |
|---|---|---|---|---|
| Short (1–2 words) | ~40% | ~35% | ~40% | 12.8% |
| Medium (3–5 words) | ~60% | ~50% | ~50% | ~20% |
| Long-tail (6+ words) | 100% | 73–87% | 73–87% | 43.5% |
AI search pushes all queries longer and more conversational. When someone asks ChatGPT or Perplexity a question, they don’t type “CRM.” They ask “what’s the best CRM for a 50-person team that needs invoicing?” The data shows that third-party content dominates these exact query patterns. If your brand isn’t being cited in press coverage, analyst reports, and community discussion about these specific, detailed questions—you’re invisible to AI search.
Higher Stakes = More Third-Party Dominance
The more commercially valuable a keyword (higher CPC), the more likely third-party content is to dominate. The keywords brands spend the most on are the ones they’re losing organically.
| CPC Range (SaaS Platforms) | Reddit Win Rate |
|---|---|
| $15–$20 | 45.9% |
| $20–$30 | 54.8% |
| $30–$50 | 63.7% |
| $50+ | 67.3% |
The Antidote: Content Strategy Wins
The single most important finding in the entire study: UCaaS vendors (RingCentral, Nextiva, Dialpad, Five9) fight back and win. Their Reddit Threat Index is 22/100 vs. Sales Tech’s 93/100. Same algorithm, same Reddit communities—the difference is sustained investment in educational, informational content.
UCaaS vendors built glossaries, comparison guides, “how to choose” articles, and educational explainers. Sales Tech vendors relied on product pages and gated PDFs. The vendors who invested in the kind of helpful, educational content that journalists and analysts also produce? They beat Reddit. For PR professionals: this validates that thought leadership, bylined articles, media-placed explainers, and educational earned content aren’t just brand-building—they’re search strategy.
Tool Stacks for the New PR Landscape
If the data tells us that third-party content dominance is accelerating, PR teams need a different toolkit than they had in 2020. Here’s what the landscape looks like in 2026.
Search Intelligence & Monitoring
You can’t influence what you can’t see. The first investment is visibility: knowing where third-party conversations are happening about your brand, your category, and your competitors—in real time.
Tools that track SERP positions for your key terms and show you which third-party sources are ranking. You need to know when a Reddit thread, news article, or community post enters the top 10 for your category terms. Traditional media monitoring covers press mentions; you need search monitoring that covers organic rankings.
LLM Citation Tracking
A new category that barely existed 18 months ago. Tools that monitor how AI search systems (ChatGPT, Perplexity, Google AI Overviews, Claude) cite and reference your brand, your competitors, and the third-party sources in your space.
When an LLM answers “what’s the best CRM for small businesses?”—does it mention your brand? Where is it pulling that answer from? If the answer traces back to a Reddit thread or a review site, that’s a PR signal, not a marketing signal. You need to know what sources LLMs are treating as authoritative in your category.
Community Listening
Foundation’s data shows that 5 subreddits drive over 1.1M monthly searches in B2B SaaS. Your category has its own version of this. Community listening tools help you identify the specific forums, threads, and conversations that search engines and LLMs are elevating as authoritative sources.
Content Strategy & Distribution
The UCaaS vendors that beat Reddit did it with content strategy—glossaries, comparison guides, educational explainers. PR teams need tools that help them plan, create, and distribute educational content at the category level, not just brand-level messaging.
Press Centres & LLM Attribution
If third-party content dominates search and AI citations, every brand’s press centre becomes a critical piece of infrastructure. Most are not built for this reality.
The Current Problem
Most corporate press centres are designed for human journalists: a list of press releases, some exec headshots, maybe a media enquiry form. They’re rarely optimized for search engines, let alone for the AI systems that are increasingly synthesizing brand information from across the web.
LLMs pull from indexed web content to construct answers about your brand. If your key facts, statistics, leadership positions, and product information live in PDFs, behind logins, or on pages with poor semantic structure—AI systems will reconstruct your story from whatever is accessible. Usually that means Reddit threads, review sites, and competitor content.
What an LLM-Optimized Press Centre Looks Like
It’s structured data, not brochureware. Think of it as making your brand’s story machine-readable:
Structured facts: Key statistics, milestones, and positioning statements in clean HTML (not PDFs) with schema markup. Quotable summaries: Short, authoritative statements that LLMs can directly cite. Category context: Educational content that positions your brand within its market category. Accessible data: Research findings, survey results, and benchmarks in formats AI can parse. Fresh content: Regular updates that signal to both search engines and LLMs that this is an active, current source.
The Action Items
Audit your press centre. Can an LLM extract your key facts, recent news, and category positioning without downloading a PDF or clicking through a JavaScript-heavy page? If the answer is no, you have a PR infrastructure problem that’s costing you visibility in every AI-powered search result, every day.
Methodology
Source Research
This analysis is a reframing of Foundation Inc’s study by Ross Simmonds, published March 12, 2026 (updated March 25, 2026). Foundation’s team analysed 8,566 keywords from keyword gap reports covering every term where Reddit.com and at least one of 13 major B2B SaaS domains both hold a ranking position.
What Was Measured
For every keyword: SERP position, monthly search volume, keyword difficulty, CPC, competition density, search intent classification, and the actual ranking URL for all domains. A “win” was defined as Reddit holding a higher SERP position than every competing vendor simultaneously (“Beats All” rate).
The Domains (4 Verticals, 13 Vendors)
| Vertical | Domains | Keywords |
|---|---|---|
| Review Sites | G2, Capterra, Trustpilot | 2,447 |
| SaaS Platforms | Salesforce, HubSpot, Zoho | 5,056 |
| Sales Tech | ZoomInfo, Salesloft, Apollo.io | 236 |
| UCaaS / CCaaS | RingCentral, Nextiva, Dialpad, Five9 | 827 |
PR Reframing
The translation layer in this report reframes Foundation’s findings for communications professionals. Where the original report discusses “Reddit dominance,” we interpret through the lens of third-party content authority—since Reddit threads are, functionally, a form of earned, independent content. The PR implications (LLM citation, press centre optimization, earned media strategy) are original analysis by Lexi Mills.
Fact-Checking
Every statistic in this report was cross-referenced against the original Foundation report using Daedalus, an AI-assisted fact-checking tool. All numerical claims link to their source via footnotes.
- Foundation Inc: “Combined monthly volume across those keywords: 957,540 searches. Nearly a million searches per month where buyers encounter Reddit before they encounter any vendor.” Source
- Foundation Inc: Reddit “Beats All” rate of 50.3% (Review Sites), 53.3% (SaaS Platforms), 66.5% (Sales Tech). The 50–67% range reflects the outperformance across three of four verticals. Source
- Foundation Inc: “77% of the search volume that Reddit wins comes from keywords that have nothing to do with reviews, alternatives, or comparisons.” Source
- Foundation Inc: “At six or more words, Reddit wins 73-87% of the time in three verticals. In sales tech, it wins 100% of long-tail queries in our dataset.” Source
- Foundation Inc: Composite Reddit Threat Index scored 0–100, weighted across Beats-All rate, organic traffic share, keyword value share, and difficulty crossover. Source
- Foundation Inc: “In sales engagement and intelligence, it’s two-thirds.” Exact figure: 66.5% Beats All rate in Sales Tech. Source
What’s Coming Next
This is Issue #1. It won’t be the last. Here’s what I’m working on—and what I’m making open source along the way.
Mike King’s AI Search Research
The iPullRank AI Search Manual is one of the most important pieces of research in our space. I’m translating it for communications professionals—what it means for PR teams, earned media, and LLM visibility.
The PR Tool Stack for 2026
A complete breakdown of which SEO, monitoring, and AI tools PR people actually need—and which ones are noise. Honest recommendations, no affiliate links.
Free Frameworks & Resources
I’m releasing my frameworks, swipe files, and training materials—the stuff I’ve been hoarding for a decade. Free, searchable, and built for people who learn differently.
Why Open Source?
Because I believe agencies should share knowledge, not gatekeep it. Because the industry moves too fast for any one person to figure it out alone. And because I’ve seen what happens when smart people across PR, SEO, content, and martech actually talk to each other—everyone gets better.
Help Shape What Comes Next
Vote on which topics I should tackle next, or suggest something I’ve missed entirely.
Get the PDF & Stay Updated
Enter your email to get a downloadable PDF of this report and notifications when new issues drop.
No spam. Unsubscribe anytime. Email delivery powered by HubSpot.
Questions? Let’s Talk
This is an evolving landscape. If you’re navigating the shift from traditional PR to AI-era strategy, I’d love to hear what you’re seeing.