A Note from Kristin
A few months ago, I was presenting a marketing strategy to a client's board. Twelve months of priorities, laid out clearly: where to invest, what to build, how to grow.
We got through maybe twenty minutes before the questions started converging on one thing.
What are we doing about LLM strategy?
Not "how do we improve our SEO." Not "should we be at more events." The board wanted to know what their company's plan was for a world where buyers are researching vendors inside ChatGPT and Perplexity, and where showing up, or not showing up, is now a business problem.
That question didn't surprise me. What surprised me was how quickly everything else on the agenda became secondary to it.
Here's what I've come to believe: the boards asking that question are right to ask it. But the answer isn't a new technology investment or a technical fix. It's a positioning problem, the same positioning problem we've been helping clients solve for years, now with higher stakes and a less forgiving scoreboard.
Your messaging is no longer just being evaluated by human buyers. It's being parsed by AI systems that are actively building shortlists, making recommendations, and shaping the consideration set, often before a buyer ever visits your website.
This report is our attempt to explain what that means and what to do about it. Clearly, specifically, and without the noise.
Kristin Spiotto, Founder & CEO, Decoded Strategies
01. The Buyer Journey Has Moved Inside AI
If you run a B2B tech company, here's something worth sitting with: your next buyer is probably researching vendors right now. And there's a growing chance they're doing it by asking ChatGPT, Perplexity, or Google Gemini, not by Googling your category and reading blog posts.
This isn't a prediction. It's already happening.
94% of B2B buyers now use generative AI in their purchase process
50% of B2B software buyers start vendor research with AI chatbots
69% of buyers chose a different vendor than originally planned
Sources: Forrester State of Business Buying 2026; G2 Buyer Behavior Report 2025
That last number deserves more than a glance. One in three B2B buyers ended up purchasing from a vendor they had never heard of before, because an AI surfaced that vendor in response to their question. Zero prior brand awareness. Winning vendor.
That's a different dynamic than anything that came before it. In traditional B2B marketing, awareness and referrals were the primary paths to a shortlist. Now a company with the right positioning can get onto a shortlist it never knew it was competing for.
What buyers are actually asking
Buyers aren't asking AI abstract questions. They're asking practical ones:
- "What are the best [category] tools for [specific ICP or use case]?"
- "Compare [Company A] vs [Company B] for a Series B SaaS company"
- "What do customers say about [Company X]?"
- "Who are the top marketing agencies for cybersecurity companies?"
These are shortlist-building queries. The buyer is not browsing, they're narrowing down. And if your company doesn't appear in the answer, you're not on their list, regardless of how good your product or service actually is.
"The brands that show up in AI search are not necessarily the biggest or best-known. They are the ones with the clearest, most specific, most consistent messaging."
02. How AI Decides What to Surface
To understand why messaging matters so much in this environment, it helps to understand, at least roughly, how AI systems actually surface content.
Most AI assistants today use a combination of two mechanisms:
Parametric Knowledge
Real-Time Retrieval (RAG)
What the model "learned" during training. Built from web crawls, Wikipedia, Reddit, published content, and more. Changes slowly, with model updates.
Live search at query time. ChatGPT uses Bing, Gemini uses Google, Perplexity uses its own crawl. Updates much faster than training data.
When a buyer asks a question, the AI doesn't just search, it synthesizes. It pulls multiple chunks of content, filters out redundant information, and prioritizes sources that add something specific and clear. The technical term is "information gain." Content that says what every competitor says gets filtered as noise. Content that is specific, clearly structured, and adds a new data point gets prioritized.
The five signals that drive citation
Research from Princeton, Omniscient Digital, and others consistently identifies the same factors that increase the likelihood an AI will surface and cite your brand:
Signal
What it means in practice
Entity Clarity
The AI can consistently identify who you are, what you do, and who you serve, across your site, LinkedIn, review platforms, and third-party content. No conflicting signals.
ICP Specificity
Your messaging names a specific type of customer with specific problems. "B2B SaaS" is too broad. "Series A–B SaaS companies with 20–50 employees trying to build pipeline" is specific enough to match.
Third-Party Corroboration
Other sources, G2, Capterra, Reddit, industry publications, podcast appearances, describe you consistently and in alignment with how you describe yourself.
Evidence-Rich Content
Content that includes specific statistics, named expert quotes, and cited sources. The Princeton GEO study found that adding statistics to content boosted AI citation by 30.6%; adding quotes boosted it by 40.9%.
Answer-First Structure
Your homepage and key pages lead with a direct, clear statement of what you do and who it's for. The buyer's question gets answered in the first two sentences, not the fourth paragraph.
03. Why Weak Positioning Is Now a Visibility Problem
We've seen this for years in our work with clients: vague positioning creates friction everywhere. Sales conversations that go sideways. Websites that don't convert. Marketing campaigns that generate activity but not pipeline.
What's new is that weak positioning now has a concrete, measurable consequence in AI search: you disappear.
The three positioning failures that make brands invisible to AI
Failure 1: Category Drift
If your homepage says "revenue intelligence platform," your LinkedIn says "AI sales coach," and your G2 listing says "conversation analytics tool," you are competing in three different AI neighborhoods and winning none of them.
LLMs build understanding from patterns across multiple sources. When those sources conflict, the model either hedges ("Company X appears to offer a range of AI-powered sales tools") or omits you entirely.
If your positioning is vague, interchangeable, or constantly shifting, AI has nothing solid to anchor to.
Failure 2: Feature-First Messaging
LLMs reward content that connects a product to a job-to-be-done or a specific outcome. Generic "AI-powered platform" language gets filtered as noise.
Your homepage needs to answer the question a buyer would actually ask: "What is [Company], who is it for, and what does it do?" If the answer isn't in your first two sentences, you're not answering it.
Failure 3: ICP Ambiguity
"B2B SaaS" is not an ICP. It's a category. When your ICP is that broad, you're competing against every marketing agency, sales tool, and consulting firm in existence, and AI treats you accordingly.
When you define your ICP specifically, Series A B2B SaaS companies with 10+ AEs running outbound into mid-market, for example, buyers who match that description and ask precise questions get you as a precise answer. Niche positioning is not a limitation in AI search. It's a competitive advantage.
The Positioning Test
Open ChatGPT right now. Type: "Who are the best [your category] for [your specific ICP]?"
If you appear: How does AI describe you? Is it accurate? Does it match your positioning?
If you don't appear: Which competitors do? What do they have in common?
The answer is a diagnostic of your current positioning. Not your SEO. Not your ad spend. Your positioning.
04. What Strong Positioning Does in the AI Era
The same discipline that made positioning valuable before AI makes it essential now. The difference is that the consequences are more immediate and more binary.
It makes you findable for the right question
The companies that dominate AI search results aren't necessarily the biggest or most well-funded. They're the ones whose messaging most closely matches the language buyers use when describing their problem.
A company that has done the work to define their ICP precisely, and built their messaging around that ICP's specific language, shows up when that buyer asks their specific question. A company with generic messaging shows up nowhere or everywhere, which is functionally the same thing.
It creates a consistent signal across every surface
Omniscient Digital's analysis of 23,000+ AI citations found that for branded queries, only 23% of citations come from a brand's own website. The other 77% come from third-party sources: G2, Reddit, review sites, publications, podcasts.
That means positioning is not just a website problem. It's a whole-ecosystem problem. Strong positioning gives every surface, from a G2 profile to a podcast interview to a LinkedIn post, the same clear signal to send.
It creates real differentiation in a sea of AI-generated sameness
According to Ahrefs, 74% of new web pages now include AI-generated copy. The result is more content that increasingly sounds identical. Polished. Correct. Interchangeable.
LLMs have a built-in filter for this. Content that repeats consensus gets deprioritized. Content with a specific point of view, original data, or a distinctive perspective gets amplified. That's exactly what strong positioning creates: not just clarity, but distinctiveness that only your company can own.
It builds the kind of trust that converts
For all the attention on AI changing everything, the research is consistent on one point: buyers use AI to build shortlists, but they still rely on human judgment to make the final decision.
Forrester's 2026 data shows that 20% of buyers report less confidence after using AI because of unreliable or inconsistent information. Gartner predicts that by 2030, 75% of B2B buyers will prefer sales experiences that prioritize human interaction over AI-driven ones.
Strong positioning serves both ends of that process. It gets you on the shortlist. And when a buyer lands on your website, reads your content, or talks to your team, the clarity and consistency of your message builds the trust that closes the deal.
05. The Five Moves That Make Messaging AI-Ready
These are not technical fixes. They are strategic and messaging decisions, the same work that makes positioning strong for human buyers, extended to a new environment.
Move 1: Lock your category label and use it everywhere
Pick one clear, specific way to describe what you do and who you serve. Use that exact language on your homepage, your LinkedIn company page, your G2 profile, your Crunchbase listing, and in every piece of content you create.
The goal is a consistent signal that an AI system can anchor to. Every conflicting description is friction the model has to resolve. It usually resolves by hedging or omitting.
- Strong: "B2B messaging and positioning agency for Series A–B tech companies"
- Weak: "Strategic marketing partner" (no category, no ICP)
- Also weak: Different descriptions on different platforms
Move 2: Rewrite your homepage for the first-sentence test
Read the first sentence of your homepage hero. Does it immediately answer: What do you do? Who is it for? What's the outcome?
If someone asked ChatGPT "What is [Company]?" would your homepage answer that question in two sentences?
Answer-first structure isn't just good UX. It's how AI retrieval systems find and extract your core positioning.
Move 3: Build the corroboration layer
Your positioning is only as strong as what third parties say about you. Review platforms, publications, and community discussions are where AI systems go to validate what your website claims.
The practical moves:
- Complete and optimize every G2/Capterra profile with the same language as your website
- Drive 15–20 structured customer reviews in the next 90 days
- Earn one or two placements in "best of" listicles in your category
- Show up consistently in the communities where your buyers have conversations: specific LinkedIn groups, Slack communities, or relevant subreddits
Move 4: Add evidence to your content
The Princeton GEO study tested which content modifications most improve AI citation. The results:
Content modification
Lift in AI citation
Adding quotations (named experts)
+40.9%
Adding specific statistics
+30.6%
Citing sources inline
+27.5%
Keyword stuffing
-8.3% (worse than doing nothing)
Source: Aggarwal et al., Princeton / Georgia Tech / Allen AI / IIT Delhi, KDD 2024
Every piece of content should include at least one specific statistic or data point, one named expert or customer perspective, and one clearly attributed source. These aren't SEO tactics. They're what makes content genuinely useful, and what makes AI systems treat it as worth citing.
Move 5: Do the positioning work first
All of the moves above are built on a foundation. If the positioning itself is unclear, if you haven't made the hard decisions about your category, your ICP, your differentiation, no amount of optimization will create the clear, consistent signal that AI systems need.
Positioning clarity is not just a branding exercise anymore. It is now a prerequisite for being found by buyers who are using AI to research vendors.
The companies that win in this environment are the ones who do the work to get genuinely clear: about what they do, who they serve, and what makes them distinctly better for that specific person with that specific problem.
The Bottom Line
The rules of B2B marketing didn't change. They got more demanding.
Clarity of messaging has always mattered. Now it's measurable in a new way: in whether you show up or don't when a buyer asks an AI system who can solve their problem.
The companies that get this right aren't the ones who hire technical SEO specialists or invest in AI optimization tools. They're the ones who do the foundational work: getting clear on who they serve, what problem they solve, and why they're the right answer for that specific person.
That clarity, expressed consistently across every surface, is what creates visibility in an AI-driven world.
"Positioning isn't just about how you sound. In 2026, it's about whether you exist in the places your buyers are looking."
Want to know how your messaging performs in AI search? We run a focused AI Visibility Audit that shows you exactly where you stand, and what to do about it. Book a free 30-minute conversation at decodedstrategies.com
About Decoded
Decoded is a messaging and marketing strategy firm. We work with founders and GTM leaders at Series A–B B2B tech companies who need to get clear on what they're saying before they can grow.
Our work starts with a Brand Messaging Guide, built through live workshops with your leadership team. From there, we develop a Six-Month Marketing Roadmap that shows where to invest, what to build, and how to sequence it. Once the right foundations are in place, we stay involved through content execution.
Content without a messaging foundation is expensive noise. That's what we're here to fix.
decodedstrategies.com
