By Lora Kratchounova, CEO of Scratch Marketing + Media
The way buyers learn has changed. That shift is now changing who wins.
For the last decade, category leaders benefited from scale. They had the largest budgets, the broadest reach, and the most familiar names. If you showed up everywhere, you stayed top of mind. That playbook still matters, but it is no longer enough.
Large language models are becoming the first stop for discovery and evaluation. Buyers ask a question and get a synthesized answer. They compare options without visiting ten websites. They narrow a shortlist before your demand gen engine even has a chance to do its work.
Category Leaders Can Become the Default, Not the Definition
In this world, incumbency comes with a new risk. You can become the safe, recognized answer while the category quietly shifts underneath you. Challengers can define what “good” looks like next, then ride that definition into the market.
The fight is no longer only for attention. It is for meaning.
The Salesforce Problem (and Every Category King’s Risk)
Consider Salesforce. It redefined CRM as SaaS and built one of the most successful enterprise platforms in history. But a new class of LLM-native companies – such as Aurasell and Spiro, aren’t trying to out-Salesforce Salesforce. They are reframing the category.
Rather than positioning CRM as a system of record and workflow automation, these challengers are orienting the market around reasoning, guidance, and decision support. In other words, they are competing in what many describe as System 2 technology, software that interprets context, reasons across signals, and recommends action, rather than merely executing predefined workflows.
If Salesforce continues to present itself primarily as a System 1 platform in a System 2 world, it risks becoming the incumbent answer in LLM-driven discovery, recognized and respected, but no longer leading where the category is headed.
The New Competitive Advantage Is Meaning That Survives Summarization
When an LLM compresses your message, you do not get credit for nuance. You get credit for clarity. If your positioning collapses into generic category language, you lose. If your positioning holds its shape, you win.
That is the bar now. Your story has to survive compression, paraphrase, and side by side comparison.
How Incumbents Win Again
Incumbents can respond. But they have to lead with intent.
That starts by claiming the next category shift, before the market agrees on what it is called. Then it requires a stronger form of positioning. Not a tagline. Not a campaign. A narrative architecture that stays consistent across the assets buyers and AI systems actually reuse.
It also requires orchestration. You need an ecosystem that repeats the story in places you do not control. You need proof that stands up across channels. And you need a message that stays credible when it is reduced to an answer.
This is not an argument for chasing every AI feature. It is an argument for treating category definition as a strategic asset again. The companies that win will be the ones who teach the market how to think.
The Playbook Starts When the Theory Ends
If AI becomes the first interpreter of your brand, then category leadership depends on what the market, and the models, learn to repeat about you. That is the strategic frame. The next question is operational. What do you actually do next?
To answer that, I asked the leaders of Scratch’s practice areas to weigh in and share what they are seeing first-hand.
- How demand gen earns belief when the click disappears.
- How content stays distinct when sameness floods the market.
- How PR builds trust and specificity in the public record that buyers and AI systems rely on.
Read the following sections as one story told from three angles. The goal is the same in each case. Make your narrative durable enough to survive summarization, comparison, and repetition.
In an LLM-Driven Buying Journey, Credibility Is the New Conversion Event
By Peter Atanasoff, VP, Demand Generation
LLMs collapse the funnel. The buying journey has compressed from 70/30 to 60/40—but the fundamentals remain unchanged. What has changed is when deals are won or lost.
Here’s what the latest research from 6sense tells us: the vendor ranked first at the end of the Selection Phase goes on to win about 80% of the time. Most deals are effectively decided before sellers are ever contacted. Buyers are engaging vendors earlier—61% of the journey happens before first contact with sellers, yet decisions still hinge on Selection (first ~60%) and Validation (final ~40%).
And relationships still matter enormously. 85% of buyers report prior experience with the winning vendor, and 75% say they personally know sellers from the vendor organizations they evaluate. Preliminary consensus remains the strongest predictor of outcomes.
What does this mean for demand gen?
Buyers are outsourcing their first evaluations to LLMs where your brand is represented by summaries, not your home page or best landing page. 94% of buyers are using LLMs during their buying process—primarily to compare vendor offerings, evaluate proposals, and analyze stakeholder input. They are using multiple sources both owned and not owned. This shifts demand gen’s job from clicks to proof because LLMs reward consensus and sourcing.
Credibility is shaped by outside channels (reviews, analysts, tech communities and mentions) making demand gen and PR complements of each other in the visibility landscape. Content is king and can no longer be managed in silos.
Because LLMs are also being used for fact-checking, any mismatched messaging (narrative drift, legacy content) will result in an inauthentic story or at least LLM confusion. Driving stronger alignment across the entire journey, campaigns, and thought leadership are key to LLMs getting your story right. Again, the entire GTM motion needs to be in sync.
The Clear Mandate for Revenue Teams:
- Invest in capabilities to sense, interpret, and act on buying group signals early
- Create and surface self-serve, high-value content that answers Selection Phase questions
- Consider adding answer-led journeys to your web properties
- Equip sellers and CSMs to build preference outside of active cycles
- Measure success by shortlist placement and win rate—not raw lead counts
In an LLM-driven buying journey, credibility is becoming the new conversion event.
AI Made Content Cheap. It Also Made Content Boring.
By Richard Hostler, Head of Content Strategy
The internet is filling up with “good enough.” It looks polished. It reads clean. It says the right things. It also sounds like everyone else.
That creates a strange dynamic. Brands are producing more content than ever, but they are less memorable. Buyers are skimming more, trusting less, and relying on fewer sources to decide.
The Job Is Not volume, It Is Signal
LLM powered discovery accelerates the problem. If a model can summarize ten similar posts into one answer, the winner is not the brand that wrote the most. It is the brand with the clearest point of view, the most credible authority, and the strongest proof.
That is the real role of content in this era. Content is where your market learns how you think. It is where your category definition becomes legible at scale. It is where your ideas become repeatable, by people and by machines.
Point of View Buyers Can Quote and AI Can Carry
Point of view is not hot takes. It is a stance on what is changing and what to do about it. It names tradeoffs. It creates a clear “we believe” a buyer can repeat in one sentence.
If your content never risks disagreement, it is probably saying nothing.
Authority That Has a Name, a Face, and a Track Record
In the AI era, anonymous expertise does not land the same way. Buyers want to know who is speaking and why they are credible.
Put your SMEs forward. Use named perspectives. Show the thinking behind decisions. Make your expertise traceable to people, not to a brand voice that could have been generated anywhere.
Proof That Holds Up Outside Your Website
Validation is the piece most content programs underinvest in. It is the proof your point of view works in the real world.
It can be customer evidence. It can be numbers, if you have them. It can be before and after narratives. It can be a clear explanation of what changed and why. Whatever form it takes, it answers the buyer’s quiet question. Is this true, or is it just phrased well.
Build a Content System, Not a Content Factory
Many content teams are still operating like it is 2015. They optimize for volume. They follow keyword maps. They chase the next format. They ship and move on.
That made sense when search rewarded frequency. But AI mediated discovery compresses information and averages the internet. If you are average, you disappear.
So content leaders have to shift from production to architecture. Audit your narrative the way product teams audit usability. Where do we contradict ourselves. Where do we sound generic. Where do we avoid specificity because it is harder.
Then design for distribution beyond your site. Some of your most important content is not a blog post. It is a founder memo. A strong LinkedIn post that sparks discussion. A customer story others can cite. A point of view your sales team uses because it actually helps.
We can also take a lesson from the early days of SEO, and the radical content approaches that gained short-term visibility at the cost of buyer relevance. As algorithms became more sophisticated, deeper, more human-centric content won out.
AI will not kill content. It will punish sameness. It will reward the brands that have something real to say and the discipline to say it consistently.
PR Is the Training Data for Your Category
By Anya Nelson, SVP, Comms & PR
If your narrative only exists on owned channels, you do not have a category. You have a campaign.
LLMs do not just read your website. They synthesize the broader web, looking for patterns and, most importantly, consensus. If content is what you say about yourself, PR is what the internet says about your brand. In an answer-first world, that consensus becomes the buyer’s reality.
LLMs validate claims by checking whether authoritative third-party sources agree with your positioning. That means PR cannot be measured only by awareness. It has to place verifiable proof in the domains answer engines trust.
Citation Architecture Is the New Mandate
When AI becomes the first interpreter of your brand, PR becomes less about getting seen and more about engineering belief. Buyers ask a question and get a synthesized answer before they ever see your homepage. That answer is shaped by what the market has published, repeated, and validated across sources you do not control.
When an incumbent’s story collapses into generic category language, the model is not misunderstanding you. It is reflecting a public record you failed to shape.
Build a pattern of clear, consistent third-party signals that keep your positioning intact under compression. LLMs reward consensus, specificity, and traceable authority. So lean into reputable outlets, structured sources, named experts, and proof that can be checked.
This is also why you can’t turn the PR switch off and on as you please. LLMs learn from patterns over time. One-off hits do not outweigh a steady chorus of consistent language across analysts, partners, reporters, and credible digital creators. PR has to coordinate tightly with content and demand gen so the story stays aligned everywhere buyers (and models) look.
The PR Playbook for Answer Engines
To earn visibility and trust, PR teams should prioritize the following activities.
- Dominate “bullseye” publications: Identify which outlets consistently surface in AI answers for your category. Study what they cover, who they quote, and how they frame stories. Then tailor a pitch strategy that earns repeat inclusion.
- Pitch for inclusion, not announcements: Funding rounds and product updates age fast. Inclusion in trend pieces, vendor comparisons, and “what’s next” narratives ties your brand to the market shift. Make quoting you frictionless with a tight POV, unique data, specific customer outcomes, and named experts.
- Seed category language via contributed content: Bylines and op-eds let you define the narrative frame. Write with crisp definitions, explicit tradeoffs, and durable proof points so the framework survives paraphrasing.
- Draft off established domain authority: Collaborate with industry groups, associations, and hyperscalers. Secure mentions across their blogs, resource hubs, directories, webinars, and partner pages. LLMs often treat these organizations as objective category referees due to their high authority or massive digital footprint.
- Turn analysts into your validation engine: Analysts publish the structured language models reuse. Invest in relationships beyond the standard briefing cadence. Create pathways for independent media commentary that reinforces your positioning.
- Operationalize AEO monitoring: Track how your category, your competitors, and your claims show up in AI answers. Monitor which publications drive citations and where your message drifts, then use that visibility to guide outreach and react faster to emerging narratives.
In an answer-first world, the right PR strategy and flawless execution is how you force the market—and the models—to agree on exactly what you mean. But it’s not a magic wand. Don’t try to trick the LLMs. It didn’t work with SEO and it won’t work with AEO. Be intentional about quality and long-term value.
The Bottom Line: Act Now or Let AI Define You
VMware Proved That Categories Are Built, Not Found
There is a clear precedent for this moment. When VMware introduced the software defined data center, the market was not asking for it. The category did not exist.
VMware made it real by seeding the ecosystem with shared language. Partners like Cisco, F5, and Pivotal reinforced the same story across integrations, co marketing, and field narratives. Analysts learned the frame. PR carried it. Campaigns repeated it. Over time, VMware did not just show up in that category. It shaped the category, and owned it. (And our team was lucky enough to be part of this journey.)
The same playbook applies now with one critical difference. LLMs are part of the audience.
Four Moves That Defend and Reclaim Leadership
If you are an incumbent, you do not “opt into” category definition anymore. You either lead it, or you inherit it from challengers and from what answer engines learn to repeat.
- Claim the next category early: Decide what the category becomes next, then put language around it before the market settles on a default. If you do not define the direction, AI systems will infer it from whoever does.
- Re-architect positioning at the source: Treat your core pages and foundational assets as training signals. Build a language architecture that is consistent, concrete, and hard to paraphrase into generic category noise.
- Activate the ecosystem as a force multiplier: Categories become real when partners repeat them. Align integrations, alliances, and co marketing around the same outcomes and the same words.
- Run PR, analysts, and social as one system: LLMs learn from patterns of consensus. Sustained reinforcement across credible third parties beats one off announcements every time.
What to Do in the Next 30 Days
- Week 1, define the “answer” you want to own: Write the three sentences you want an LLM to say about your category and your role in it. Stress test them for clarity and proof.
- Week 2, clean up narrative drift: Audit your top pages, your sales story, and your highest visibility content. Remove contradictions. Replace vague claims with specific, verifiable ones.
- Week 3, build proof that travels: Create a proof plan that lives beyond your site. Customer outcomes, comparisons, expert perspectives, and independent validation should support the same claims.
- Week 4, orchestrate repetition: Brief partners, analysts, and spokespeople on the same language. Then reinforce it through earned media, social, and targeted demand programs that push the proof into the market.
The Stakes Are Clear
In this era of LLM-driven buyer journeys, the companies that win will not be the ones that publish the most or shout the loudest. They will be the ones that teach the market what to believe, and give it a reason to agree.
