The Journey You Inherit

At the Chicago stop on the Spotlight on the Road tour, a room full of B2B marketing, AR, and customer engagement leaders confronted the same uncomfortable fact from five different angles: the buyer journey isn't something a vendor runs or steers anymore. It's something they inherit. Buyers are already half-decided, at least, long before they contact vendors.

There's a tug-of-war at the center of B2B marketing that's as old as the discipline itself: what you say about yourself versus what others say about you. What you say — your specs, your differentiation, your own content — still matters. It has its place. But the data and the patterns are consistent: what they say — your customers, the communities where they gather, the analysts and experts who cover your category, even your partners and employees — carries more weight than ever, especially when you're not “in the room.”

Most B2B teams, though, are still oriented toward the owned side of that ledger. It's predictable. It's legacy. Old habits die hard. And that's a problem, because in a delicious twist, the "robots" — ChatGPT, Gemini, Copilot, and the rest — prefer earned, human-derived authority. Authentic customers. Authentic experts. Real people. The fastest-adopted technology in human history (ChatGPT) has made that preference matter more than it ever has.

The stakes:

• 94 percent of B2B buyers now leverage generative AI at some point in their journey (Forrester, Buyers' Journey Survey, 2025).
• Four out of five deals are won by the buyer's own predetermined favorite — decided through "dark research" before they ever officially talk to a seller (6sense, 2025 B2B Buyer Experience Report).

Buyers are showing up more confident, more informed, and more certain of what they want than ever before. Or they're not showing up at all. Without a system to understand and orchestrate what's being said about you when you're not in the room, you're in the dark — and buyers won't be able to find you either. The new reality follows from that:

• You don't guide the buyer journey anymore. You inherit it.
• You're winning or losing the deal before the conversation starts.
• What the robots trust most is exactly what we trust most in each other: peers and experts.

The challenge for the room, and the thread that ran through every session in Chicago: Think about how to be more than a result. Think about how to be the answer.

Winning the Answer: How Trust, Reviews, and Customer Voice Shape Modern Buying

Christine Li, VP, Global Partnerships and Data Solutions at G2, made the case for why there's no richer supply of customer trust signals than peer reviews, and why that matters more than ever now that LLMs are involved.

G2's own research shows the shift in buyer behavior starkly: the share of B2B buyers who say they start their research with AI grew from 29 percent to more than half between last spring and March of this year. What that does, functionally, is compress the buyer journey. The old model — an average buyer consuming 10 pieces of content across multiple channels before reaching out to sales — is collapsing into a single prompt and a single answer. Buyers rarely click through to verify unless they distrust what they're shown. The implication is blunt: if you're not on the shortlist the AI generates, you don't exist to that buyer.

To get on that shortlist, Christine Li argued, you have to think like the machine. LLMs are optimizing for the most accurate answer, and they get there by weighing trust, authority, and volume of consensus — not by trusting what a brand says about itself, which they know is biased. That's exactly where user-generated content sites like G2 (and Reddit and several more) come in. G2 is already the No. 1 cited source for B2B software inquiries, a lead that's grown since G2's acquisition of Capterra, Software Advice, and GetApp (together known as Gartner Digital Markets).

She shared Spotlight's own case study as proof of the mechanism: Spotlight grew its G2 review count from one to 25 by sourcing reviews at its in-person events, then used Profound to measure whether that showed up in AI search. Over a 28-day window, running 1,433 prompts simulating a late-stage buyer evaluating Spotlight, G2 was cited in 68 percent of all prompts — climbing to 82 percent on ChatGPT and Gemini, and 93 percent on Google AI Mode. Every major LLM tested cited G2.

Her framework for winning the shortlist: Make your content LLM-friendly (ungated, factual, speaking the language answer engines actually parse), close the gap between the narrative you want to push and the accurate, consensus-based answer the LLM is looking for, and earn the trust signals that make that possible. For companies without an established customer base yet, the weighting still applies — trust and authority sources like G2 and Reddit carry heavy weight, but where those don't exist yet, LLMs look elsewhere, so your own content still has a job to do filling the gap. G2, Spotlight, and Profound have packaged a starter kit together to make that easier to act on quickly.

Audience questions pushed further into the mechanics: G2 employs more than 80 review specialists vetting and validating submissions, with a rejection rate around 30 percent that's climbed closer to 40–50 percent since the rise of AI-generated review bots. G2 is working toward a unified review form across G2, Capterra, and Software Advice, along with a centralized admin portal for managing reviews across all three — while keeping each brand's distinct strength intact (Capterra skews SME, G2 skews mid-market and up). On amplification, G2 offers content syndication tools and marketplace integrations (AWS, Azure). Her advice for where to start Monday morning: check your G2/Sell G2 login, since AEO citation tracking through the Profound partnership is already built into the back end. Know where you stand today before deciding where to go next.

Customer Advocacy & Influence Orchestration

In a fireside conversation, Bianca Del Vecchio, Managing Consultant at Spotlight (via Captivate Collective) and a marketer with more than 10 years spent turning customers into advocates, walked through what customer advocacy actually means today, in the AI era.

Customer advocacy gets talked about mostly in terms of outputs: case studies, peer reviews, references. Those still matter. But the discipline is really about working with your most engaged customers — your fans — and creating value for them, not just extracting it. Reciprocity is the word Bianca kept returning to. Storytelling remains at the center of that work, because people connect with stories, and — in a useful coincidence — so does AI. The difference is that machines synthesize and aggregate those signals before a human ever gets the chance to make an emotional connection to them, which means brands have to actively manage how their narrative shows up rather than assume it will land the way they intend.

Several years ago, Gartner moved away from vendor-supplied references for its Magic Quadrant process in favor of Gartner Peer Insights. That was a signal that analyst relations and customer engagement were destined to converge. Both functions are pillars of trust: analysts shape expert perception, customers provide peer proof, and both buyers and AI are drawing on the same sources to form an opinion. That's the idea behind Influence Orchestration — not running separate analyst and customer programs, but recognizing they feed the same engine. Fragmented signals produce an incoherent, aggregated story; aligned signals produce a coherent one.

That shift changes the central question practitioners should be asking. It used to be "how do we break through the noise?" Now it's "are we in the answer?" and how do you become the first choice before anyone even talks to you? Collecting customer reviews isn't enough on its own. "Activity does not equal alignment," Bianca said. Cycling through the same customers for quarterly reviews is transactional, and eventually you run out of customers to ask.

Her framework for where the discipline is headed is Precision Advocacy. It’s the latest stage in an evolution that ran from legacy reference programs, to scaled advocacy enabled by tooling, to lifecycle advocacy embedded in the customer journey, and now to a surgical, AI-enabled approach. It isn't about maximizing the volume of stories or reviews. It's about understanding what customers are actually saying about you in LLMs versus your competitors, finding the gaps, and filling them deliberately. Her advice to start: audit every channel where customer voice shows up, check whether those communities and conversations are actually accessible to LLMs (not gated), and see whether the story being told matches the one you want to be heard.

For those outside customer engagement wondering how to get involved: start connecting the dots internally. Find out who owns analyst relations, who owns customer engagement, and get in a room together around a shared question: What is being said about us in the LLMs, and is that how we want to show up? On the question of what makes storytelling powerful, Bianca's answer was simple: focus on the human piece. Business metrics matter, but the moments people actually remember and repeat are the ones that show how technology made someone's day-to-day easier. 

Questions from the audience sparked more compelling discussions:

Q: What’s with Gartner Peer Insights' inconsistencies [across categories and in analyst usage]? 

A: It’s a challenge, to be sure, but the trajectory is toward more adoption by analysts and more consistency in the taxonomy. Gartner selling its Digital Markets assets to G2 allows Gartner to put more resources into Peer Insights. 

Q: For early-stage vendors in undefined categories, how can they build authority in customer reviews without an established category in the review sites?

A: Look for the pain-point conversations happening in tangential communities and reviews. Use those to illustrate the need for the new category.

Be the Answer. But Can Your Team Deliver It?

Fred Faulkner, SVP of Marketing at McFadyen Digital and founder of Get AI Literate, brought the conversation back to the people and processes inside a marketing organization that have to execute on everything the morning had covered so far.

He opened with a personal example of the stakes. Scrolling LinkedIn, he found his own company tagged in a post ranking "top e-commerce software and systems integrators" — one that categorized McFadyen Digital strictly as a marketplace player, a positioning the company has spent years actively moving away from. It raised real questions: Do you correct that in the comments? Do you publish content elsewhere to cover current capabilities? All of the above. The deeper issue is a double-edged sword: AI-driven visibility is exactly what you want, but only when it reflects where your business actually is today, not where it used to be. As companies evolve their go-to-market strategy, the citations and sources that authority is built on have to evolve with them, and most organizations don't have a plan for managing that shift.

To make the point concrete, Fred ran the room through a live exercise: five short customer-review excerpts, asking the audience to identify each one as human-written or AI-generated. The results landed close to an even split on nearly every example, with no clear consensus — including one case where a well-written human review fooled almost the whole room into guessing AI. That's the trust problem in miniature: content volume is exploding because AI makes it trivial to produce (a 10-bullet list can become a "10-page white paper" that reads as more authoritative simply by virtue of its format), and distinguishing authentic voice from synthetic content is only getting harder.

His broader argument is that AI fluency is no longer a marketing-department problem; it's an organization-wide one. Anyone publishing content publicly, on LinkedIn or elsewhere, is now part of what shapes "the answer" for their company. Yet most organizations, in his experience, don't have a formal AI literacy program beyond an ad hoc lunch-and-learn, and leadership tends to overestimate how AI-literate their teams actually are. He proposed seven pillars of AI fluency — covering when to use AI versus not, critical thinking, measuring output quality and improvement, and governance guardrails — and argued companies need a real diagnostic, not just an assumption that people can "use ChatGPT."

On strategy, his key point was that content is no longer about volume; it's about authority. Every function in the business, not just marketing, is now producing material (case studies, reviews, commentary) that becomes part of the answer, and someone needs to coordinate that. He also challenged the room's attachment to the MQL as a metric, arguing it's effectively dead given how much of the buying journey happens before first contact, even though many organizations still use it. And he named the gating dilemma directly: withholding strong research or IP behind a form protects lead generation, but ungated content is what LLMs can actually index and surface. It’s a real strategic tradeoff, and not an easy call.

His closing challenge to the room: talk to whoever owns your organization's content strategy and ask how you're adapting beyond traditional SEO, and whether you're measuring it. And personally audit your own public footprint — presentations, press, reports — for accuracy and authority, since you're now part of the answer too. In Q&A, he pointed to concrete warning signs that an organization isn't as AI-literate as leadership assumes: a leadership team pushing AI adoption without being able to articulate a clear strategy, and the absence of any measurement tied to AI-driven efficiency gains (he cited an example of a company burning a full year's AI token budget in four months with no plan to measure whether it was working). He also argued for AI-fluency training tailored by seniority level rather than one-size-fits-all, and for being deliberate about which tasks get delegated to AI — synthesizing and summarizing large volumes of material is a strong use case, provided a human still validates the output, since accountability for the result doesn't go away just because AI produced the draft.

Winning the Selection Phase

Kerry Cunningham, Head of Research and Thought Leadership at 6sense and a longtime B2B researcher formerly of Forrester and SiriusDecisions, closed the Chicago program with a case for why most B2B revenue systems are built for a world that no longer exists.

His argument starts with a historical point: the MQL, BDR sequences, and funnel-stage thinking that still dominate B2B marketing were built roughly 20 years ago, at the dawn of the internet era, when buyers genuinely had to talk to a seller to get basic information. That hasn't been true in most established categories for a long time. Buying groups today average around 10 people — a number that Kerry noted hasn't actually grown over the past decade, contrary to the popular narrative, because it's driven by internal process and the human instinct not to be the one who gets blamed if a purchase goes badly, not by the market.

The bigger point: 75 to 80 percent of B2B purchasing happens inside already-established categories, where every buyer enters the process with prior experience and opinions already formed. Only 20 to 25 percent of purchasing happens in genuinely new categories requiring real discovery. The average B2B buying cycle runs about 10 months in North America, 11 elsewhere. At any given moment, roughly 60 percent of a company's ideal customer profile isn't in market at all, about 30 percent is in market and persuadable, and only around 6 percent is in the late validation stage.

The shortlist dynamics are the crux of it: buyers put four or five vendors on their shortlist on day one, based entirely on prior experience and reputation, and 95 percent of purchases end up coming from that day-one list. First contact with a seller happens, on average, 61 percent of the way through the buying journey. Ninety-four percent of buyers say they'd already ranked their shortlist by preference before that first conversation. What's happening in the gap is internal consensus-building — and per research from the B2B Institute at LinkedIn, 40 percent of buying processes fail simply because the buying group can't agree on requirements or on which vendor should top the list. Sellers report initiating contact 78 percent of the time, but from the buyer's side, outreach only gets a response once their internal group has reached consensus — the timing of a BDR's email has essentially nothing to do with when a buyer responds. The vendor at the top of the shortlist wins the deal 77 percent of the time; sellers only flip the final decision in 23 percent of deals, though they shift stated preference in 42 percent — meaning sellers matter more as a long-term brand asset than as a short-term lever.

Kerry's practical framework reframes the journey into a selection phase (largely invisible to sellers, decided through peers, analysts, and personal experience) and a validation phase (where traditional sales and marketing activity is concentrated). Losing the selection phase makes winning the validation phase nearly impossible. His recommended posture is to "aggressively enable" rather than aggressively pursue meetings — identify everyone in the buying group who needs to be influenced and deliver them the best information and experience possible, without expecting or measuring success by whether they respond.

On LLMs specifically, his data complicates the popular narrative: buying journeys have compressed, but it isn't primarily because of LLM usage — buyers who don't use LLMs actually report shorter journeys in 6sense's data. The bigger driver is that vendor products themselves now embed AI, which adds a new evaluation dimension (and often pulls in new stakeholders, like security or AI governance teams) even for experienced buyers. Where LLMs are used, it's more often in the middle of the journey — to support and confirm preferences buyers already hold — than at the very beginning to discover new vendors.

The strategic implication: concentrate demand-gen and ABM spend on the roughly 30 percent of accounts that are in-market and persuadable early, and shift meaningfully more investment toward branded influence of the 60 percent not yet in market, since that's where shortlist preference actually gets formed. Signals like leadership changes or M&A are useful for beginning to nurture a relationship, not for triggering an immediate sales push — revenue impact from those events typically takes 12 to 18 months to materialize. On the value of events specifically, Kerry argued brand presence at conferences matters because buyers who aren't in-market are unlikely to visit your website and may never see updated positioning otherwise; showing up is as much about signaling community membership, and motivating internal champions to advocate for you, as it is about lead generation.

Closing Thoughts

Chicago's morning built, session by session, toward the same conclusion we opened with: You don't guide the buyer journey anymore, you inherit it — and the sources the robots trust are the same ones your buyers trust.

• Bianca showed what precision advocacy looks like in practice: not more volume, but a surgical audit of where customer voice already lives and where the gaps are.
• Christine Li proved the mechanism with Spotlight's own numbers — a G2 presence built from 1 to 25 reviews, cited in the overwhelming majority of prompts across every major LLM tested.
• Fred brought it back to the people and processes that have to execute on all of it, and made the case that AI fluency is now everyone's job, not just marketing's.
• Kerry grounded the whole morning in data: the deal is largely decided before a seller is ever contacted, and the teams that win are the ones investing in influence long before a buyer is in market.

My challenge to the room stands: Be more than a result. Be the answer. The brands that win from here won't just show up in an AI-generated response — they'll be the ones that response is built around, because their customers vouch for them, their reviews hold up under scrutiny, their content is built for the questions buyers are actually asking, and their teams are fluent enough to keep it that way as the tools keep changing. That kind of influence isn't manufactured. It's earned, one precise, well-orchestrated signal at a time.

Spotlight on the Road 2026

San Francisco: April 1 — recap here

New York City: May 6 — recap here

Seattle: May 13 — recap here

Boston: June 3 — recap here

Chicago: June 24

Austin: June 26 - recap here

Spotlight Summit 2026

Influence + Advocacy + Visibility = Trust

September 14–16 | Kansas City, Missouri

Registration is open!