Learnings From B2C For B2B Marketing Leaders
Seventy percent of marketing leaders know their programs need to change for how AI has rewired buyer behavior. Yet, most of them aren't changing anything.
That's the headline finding from Iterable's 2026 Customer Engagement Report, built around what they're calling the Apex Consumer, customers sophisticated enough to game the systems marketers built to influence them. Cart abandonment to trigger discount codes. Free trials started with the cancellation already scheduled. Service-hopping based on whatever promotion happens to be live.
If you're running a B2B SaaS company, none of this feels like your problem. I mean, you're not fighting discount codes.
Want more from The CMO?
Sign up for a free membership to complete reading this article:
Have an account? Log In
You're up against something surreptitious, and in some ways more consequential. Your buyer is forming an opinion about your company before they ever talk to your sales team, prior to filling out a form, and often before they visit your website at all.
An AI tool did the research on your behalf. The only question that matters is whether what it found made you look good.
Where Your Buyer Is Doing Their Research
Jason Ing, CMO at Typeface, spends his days building marketing tools for enterprise teams navigating this exact shift. He points to a concept making the rounds in AI circles right now, and he refers to it as the context graph.
The idea is that an AI system needs persistent memory about who you are and what you do, or it starts inventing answers without the essential, relevant information. Ing shares what happens, as result:
A lot of their AI output is deviating from what they intended because there is no memory or the AI doesn't have that context.
I'd push this further. This isn't only a content problem.
When a prospect asks ChatGPT or Gemini to summarize what your company does, compare you against a competitor, or recommend a vendor in your category, the model pulls from whatever context exists about your brand across the internet. Thin or generic context gives the model nothing worth retrieving, so you don't show up.
Ing's team built what he calls a brand agent into their product to address this directly. "Everything starts with being trained on your brand," he said.
Companies that have already defined what they stand for, in language specific enough for a model to retrieve and repeat, show up in these conversations. Companies running on generic positioning don't.
One Agency Tested This on Itself
Gwen Hammes, co-CEO of ChroMetrics, didn't wait around for someone else to write the playbook. Her agency ran the experiment on themselves first.
Over the past year, as Google search started shifting and AEO and GEO (answer engine optimization and generative engine optimization) became unavoidable topics, ChroMetrics restructured its own site.
They improved data structured, drafted clearer summaries, and created more visible information about who was writing their content. All of these aspects signal trust to search engines, LLMs, and people alike.
The result netted an 80 percent increase in referrals from large language models.
What surprised Hammes wasn't the volume of visitors. It was who showed up.
We're seeing, we obviously have the referral source, whether it's ChatGPT or Claude, and just actually seeing those come in much more so. And funnily enough, it's from big clients and enterprises.
These buyers had already done their homework. They'd asked a model about agencies in ChroMetrics' category, gotten an answer that included ChroMetrics, and arrived ready to talk.
Compare this to typical website traffic, where, as Hammes put it, "a lot of times we find they would look at our client list or our case studies" before deciding whether to engage at all. LLM referrals skip that evaluation step entirely. They arrive qualified.
If you're a CMO reading this, here's the version of the Apex Consumer that applies to you: your buyer has already done the comparison shopping. The only open question is whether your brand made the list.
What The B2C Data Highlights for B2B
The Apex Consumer research documents something B2C marketers are living in their metrics right now: consumers aren't leaving because brands failed to reach them.
They're leaving because the marketing got predictable enough to reverse-engineer. Three in five have already abandoned a platform over irrelevant content. More than half stay loyal to brands they trust for over a decade. The gap between those two outcomes isn't volume or channel coverage. It's whether the brand gave them something specific enough to hold onto.
B2B buyers are running the same calculation, just earlier in the funnel. They're using AI tools to pre-evaluate vendors before sales ever enters the picture, which means the quality of your brand signal, how clearly and specifically you've defined what you stand for, now determines whether you're in the conversation at all.
The report also surfaces a stat that should land differently for marketing leaders than it does for practitioners: 22% of marketers fear AI is already flattening brand differentiation. If everyone is using the same tools to generate the same content at scale, the brands that built something specific before the flood are the ones that still stand out in it.
Knowing Isn't the Hard Part
Priya Gill, CMO at Iterable, built the research behind the Apex Consumer framing. She's also the one who quantified the gap between knowing and doing.
Seventy percent of marketing leaders, by her team's count, understand their programs need to change. Most haven't acted, because they've decided the risk of changing course outweighs the risk of standing still.
I find that calculation backwards, and so does Gill. Her team operates on 80 percent confidence rather than waiting for the full picture.
Indecision at this pace carries its own cost. In their first few months running this way, her team freed up hundreds of thousands of dollars just by auditing redundant tools already sitting in their stack, which created both budget and the credibility to keep moving.
I hear a version of this gap in nearly every conversation I have with marketing leaders right now. You can describe the problem accurately. What's missing is the decision to move while the picture is still incomplete, because by the time it's complete, the window that mattered has usually closed.
What AI Is Amplifying
Liza Adams, an AI marketing transformation advisor, gave me the clearest way I've heard to think about why some brands show up cleanly in AI search and others don't.
"AI just amplifies what's there," she told me. "If we're an amazing company, an amazing human being, it amplifies that. And if we're otherwise, it will amplify that too."
Her advice to marketers anxious about cracking the algorithm: stop trying to crack the algorithm. "Forget the algorithm. Stop chasing the algorithm. Be an amazing brand."
That advice sounds almost too simple sitting next to Ing's context graph explanation or Hammes' 80 percent jump. It isn't. The context graph is the infrastructure. What gets amplified through that infrastructure is whatever your brand has actually built. So, things like specific positioning, a validated point of view, along with content that says something a model can retrieve and repeat with confidence.
Generic messaging produced at volume gives a model nothing to work with, regardless of how sophisticated your technical setup is.
Start Where You Are
You don't need an enterprise transformation budget or a six-month roadmap to start this work. ChroMetrics ran its experiment with existing resources. The brand agent concept Ing describes is fundamentally about clarity of positioning, work that's overdue in most marketing organizations regardless of what AI does with it.
Here's where I'd start if I were you. Ask ChatGPT, Claude, and Gemini what they know about your company and how you compare to your top three competitors.
Read what comes back with a critical eye. Then, look at your own published content and ask whether it gives a model anything specific to retrieve, or whether it reads like everyone else's in your category.
Pick one gap you can close this quarter. Ask your team if they agree with this finding, and encourage them to do the same exercise. This is your starting point.
You're one of the 70 percent who already knows something needs to change. The research exists, the case studies exist, and the mechanism behind all of this is increasingly well understood. The only thing left is deciding to move before you have the complete picture, while being early still means something.
Start Here, This Quarter
Open ChatGPT, Claude, and Gemini. Ask each one what your company does, who you compete with, and why a buyer in your category should choose you. Read what comes back without flinching. That output is a reasonable proxy for what your buyers are seeing before they ever talk to your team.
If the answers are thin, generic, or indistinguishable from your competitors' positioning, that's the gap. Fix the published content that feeds those models: sharpen your point of view, add structured data, make sure the people behind your thinking are visible as authors. ChroMetrics ran this experiment on itself and saw an 80 percent jump in LLM referrals within a year, with enterprise clients arriving already qualified.
You already know something needs to change. The B2C data shows what happens when brands wait too long to act on that knowledge. The B2B version of that reckoning is already underway, it's just happening upstream of your metrics, in the AI conversations your buyers are having before they reach you.
What’s Next?
For the latest insights, join us at the CMO Club for interviews with marketing leaders, and resources that can help you drive growth in your business.
