Table of contents
Accurate B2B buyer journey measurement has become one of the most pressing challenges facing marketers today. It’s hitting everyone from CMOs defending budgets in the boardroom to analysts staring at dashboards that no longer add up.
So, how much of the buyer journey is digital? Well, virtually all of it. And how much of it is actually visible to your analytics stack? Far less than you think.
The tools we’ve relied on for the past decade are losing their grip on buyer behavior. Metrics like organic traffic volumes, last-click attribution, and form fill conversion rates – once the cornerstones of B2B marketing ROI reporting – are increasingly unreliable signals in a world shaped by AI-driven research and zero-click search.
AI-powered research tools, zero-click search journeys, and tightening consent legislation have created a perfect storm of attribution gaps, and the marketers who are winning are the ones who’ve stopped pretending the data is complete.
This isn’t a technical glitch that a GA4 configuration fix will solve. The traditional demand gen model (drive traffic, capture leads, nurture, convert) is structurally broken. Understanding why, and building a smarter way forward, starts with facing the full scale of what’s gone dark.
That is, the buyer behaviors that are either entirely untrackable, or that can no longer be tracked with the accuracy marketers have traditionally relied upon.
The ‘new’ visibility crisis in the B2B buyer journey
For decades, someone saying, “organic traffic is down” would be a signal that something had gone wrong. A Google penalty, a competitor surge, a content quality issue – you get the idea. Today, it’s just as likely to point towards a fundamental shift in buyer behavior.
The rise of ‘zero-click’ buyer behavior
The introduction of AI Overviews into search results has had a dramatic effect on click-through behavior. Research by BrightEdge in May 2025 found that impressions increased nearly 50% year-over-year while CTR declined close to 30% after AI Overviews launched – a pattern visible across sectors and audience types.
HubSpot, one of the most SEO-sophisticated brands in B2B, saw its organic traffic fall from 13.5M visits in November 2024 to under 7M by December, an 80% year-over-year decline.
Here’s the kicker: this dramatic falloff in Hubspot’s organic clicks occurred despite the brand remaining highly visible in search results and, notably, being cited in LLMs more than any other CRM. Rankings up. Traffic down. Citations growing. It’s a paradox that exposes just how inadequate our legacy measurement frameworks have become.
Modern B2B buyer journeys & the ‘Great Compression’
Layered on top of this all is a broader compression of the research and evaluation stages of the B2B buying journey.
According to Magenta Associates 2025 research, 66% of UK decision-makers use AI tools like ChatGPT, Copilot, and Perplexity to support vendor or supplier research. Buyers ask AI tools to synthesize vendor comparisons, shortlist options, and answer procurement questions that previously required multiple site visits, whitepaper downloads, and form fills. Notice how they’re all touchpoints Marketing could track.
That synthesis now happens in a single ChatGPT session, leaving no measurable fingerprint. Buyers are overwhelmingly satisfied with this new approach: 90% of respondents said they trust AI recommendations, while 85% say they’ve discovered a new vendor or product through AI-assisted research.
Here’s the important reframe: none of this is entirely new. We’re being forced to confront a problem that never really went away – the fuzzy, imprecise, incomplete nature of B2B buyer journey analytics.
Why the modern B2B buyer journey has gone ‘dark’
The modern B2B buyer journey was already slipping through the cracks of traditional analytics long before ChatGPT ever entered the picture.
The channels where real buying decisions form – peer conversations, internal Slack threads, forwarded LinkedIn posts, WhatsApp recommendations – have always been invisible to tracking tools. We just had enough measurable activity to paper over the gaps.
“In B2B, most of the conversations that actually matter happen well before anyone goes near your website. Partner introductions, channel relationships, a recommendation from someone they trust over a pint. These things shape buying decisions months before a form gets filled in. Joining the dots between all of that and your hard data is possible. But it requires a level of measurement maturity that, honestly, most marketing teams aren’t going to get to without some specialist help.”
Understanding ‘dark’ B2B buyer behavior
Dark social is the structural reality underpinning this invisibility and drives a large proportion of B2B content sharing through private channels.
Research by SparkToro suggests 100% of tracked visits from TikTok, Slack, Discord, Mastodon, and WhatsApp are marked as Direct, and also found evidence this is happening on LinkedIn and Facebook, too. The Slack DM where a colleague pastes an article link, the WhatsApp group where a CTO asks for vendor recommendations, the email thread where procurement teams debate shortlist criteria… all of these buyer journey interactions are hidden from our view.
GenAI tools are also increasingly stripping out UTM parameters from links offered as sources in answers, usually for the purposes of privacy. Consequently, none of these interactions trigger a UTM parameter. None of them show up in your channel attribution report.
The buying committee dynamic only makes this worse. B2B purchases now typically involve 10 or more. The person who first discovers your brand, shares your content internally, or champions your solution is rarely the person who fills out the contact form.
What eventually becomes a ‘lead’ in your CRM is the final visible moment of a journey that has been unfolding, invisibly, for weeks or months beforehand. By the time you can see the buyer, they’ve already largely made up their mind.
“Dark journeys aren’t a problem to ‘fix’; they’re a reality to design for. Most of the work happens out of sight: buyers researching quietly, sense-checking options, and building internal alignment across the buying group. If we want to influence that process, we have to stop putting friction in the way – fewer forms, fewer forced hand-raisers, and more genuinely useful self-serve information that’s easy to access and easy to share internally. That’s how you earn a place on the shortlist.”
TrustRadius’ 2024 B2B Buying Disconnect Report captures this pattern clearly: 78% of buyers selected products they had heard of before they even started their formal research process. The shortlist doesn’t form during the evaluation stage. It forms in conversations your analytics will never see.
Many buyers now actively avoid direct contact with prospective suppliers and vendors during the research process. According to Gartner research in September 2024, 61% of B2B buyers prefer a rep-free buying experience.
“Many B2B buyers feel overwhelmed and frustrated by the outreach they receive from sellers and the seller’s organization. Bad prospecting actively damages relationships with potential customers.”
Cookie consent has put blinkers on your analytics
Cookie consent legislation has added another layer of opacity. GDPR and CCPA compliance means buyers can research extensively across multiple sessions and touchpoints, then grant consent late in their journey – at which point GA4 has already lost the original traffic source, logging it as Direct.
Additionally, GDPR and CCPA compliance has fractured cross-domain tracking, making it harder to stitch together a coherent journey from first touch to closed deal. This isn’t a new problem. It’s an old problem that has become impossible to defer.
“Clients query this all the time. They see a spike in Direct or Unassigned traffic in analytics and assume something’s broken or out of kilter. Nine times out of ten, it’s not. It’s just the growing reality of visitors whose full journey isn’t being captured. Maybe they didn’t accept cookie consent. Maybe they did, but later in the journey. Either way, the result is the same. You’ve got a gap in your data and there’s no way to backfill it.”
The distracted buyer: attention, indecision, and the hidden cost of inertia
There is another dimension to the dark buyer journey that marketers rarely account for: buyers themselves are becoming harder to reach, to hold, and more likely to do nothing at all. Even when intent signals are strong, the modern buyer is battling a fundamental attention deficit.
Gloria Mark, Chancellor’s Professor of Informatics at UC Irvine, has been studying how people manage attention at work since 2003. Her findings are striking: in her earliest studies, knowledge workers switched tasks on average every three minutes. By 2021, that figure had dropped to just 47 seconds. And after every interruption, it takes an average of 23 minutes and 15 seconds to fully regain focus.
The collapse of workplace attention spans
| Year | Task-switching frequency | Focus recovery time |
|---|---|---|
| 2004 | Every 3 minutes | 23 mins 15 secs |
| 2021 | Every 47 seconds | 23 mins 15 secs |
This matters for B2B marketers because the buying journey itself is a series of interrupted micro-sessions. A buyer reads half your white paper, gets pulled into a meeting, and never returns. They start evaluating you, get distracted by a competing priority, and pick up the process weeks later with different questions. Your analytics see fragments of what’s actually happening; the buyer’s experience is one long, interrupted journey.
More troublingly, distraction often leads not to a competitor win – but to no decision at all. A large-scale study of more than 2.5 million recorded sales conversations found that between 40–60% of deals today are lost not to a competitor, but to buyers who expressed a clear intent to purchase and then simply failed to act.
GenAI is the ‘fly in the ointment’ for buyer journey attribution
If dark social is the attribution problem marketers could ignore, genAI is the mosquito that keeps you up at night.
The scale of adoption alone demands attention: 6sense’s 2025 Buyer Experience Report found that 94% of B2B buyers now use LLMs during their purchasing process. That’s not a fringe behavior. It’s the dominant research mode, and it’s almost entirely invisible to marketing analytics.
A typical AI-driven B2B buying journey
- A buyer identifies a business challenge and opens ChatGPT to research their options
- They ask the AI to compare vendors, explain category differences, and recommend a starting shortlist
- ChatGPT returns a synthesized answer. No click-through. No UTM parameter. No trackable fingerprint
- The buyer navigates directly to the websites of the vendors the AI recommended
- GA4 classifies these visits as Direct traffic – indistinguishable from someone who typed the URL from memory
- The most informed, highest-intent session in that buyer’s research journey is invisible to your attribution model
Vercel: a hidden B2B buyer journey case study
The scale of this misattribution is significant. Vercel tracked this pattern directly: ChatGPT grew from less than 1% to 10% of all new signups in just six months, with the CEO sharing the data publicly.
That’s a meaningful acquisition channel that would have been almost entirely invisible to standard attribution models buried in Direct, logged as Unassigned, or credited to the wrong touchpoint entirely.
The genAI analytics gap
The absence of reliable interaction data from generative AI tools stands in sharp contrast to the measurement sophistication that exists for search, social, and paid channels.
Marketers have spent years building attribution models around platforms that were designed to be measurable. AI tools aren’t. The interaction happens in a closed environment, the synthesis happens in real time, and the resulting behavior – a direct website visit, a brand search, a sales enquiry – arrives stripped of its origin story.
Some vendors are now offering AI monitoring tools that use synthetic simulations to track brand mentions and citations across LLM platforms.
These have genuine value for understanding share of voice in AI responses, but it’s important to remember they measure AI outputs, not actual buyer behavior. A brand cited in a ChatGPT response isn’t the same as a buyer reading that citation and acting on it. The gap between monitored AI presence and real purchase influence remains unmeasured.
The conclusion for marketing teams is uncomfortable but necessary: there are ways to reduce attribution gaps (better GA4 channel grouping, server-side tracking, AI referrer domain rules, etc.) but they won’t make the problem go away entirely. The goal is to get closer to reality, not to pretend you’ve achieved full visibility.
AI tools like ChatGPT, Perplexity, Claude, and Gemini are driving a fast-growing share of B2B web traffic. But much of it hides as ‘Direct’ in GA4 or isn’t tracked at all.
To get an accurate picture of audience engagement with your brand, every B2B marketing team needs to know how to track AI referral traffic as a core analytics competency. Read our guide on tracking AI traffic.
Accepting ‘dark intent’ – understanding what you can & can’t track
The most productive shift a B2B marketing team can make right now is to stop treating dark intent as a measurement failure and start treating it as a known variable.
You can’t track what happens in a buyer’s ChatGPT session, their internal Slack channel, or the peer conversation that first put your name on their shortlist. What you can do is map where those dark segments exist and build around them intelligently.
Is your brand on buyers’ Day One shortlist?
Dark intent is buyer activity that influences purchasing decisions but cannot be observed through digital analytics.
It’s not an anomaly. In fact, in B2B, it’s most of the buyer journey. 6sense’s 2025 Buyer Experience Report found that 95% of B2B buyers purchase from their Day One shortlist – up from 85% the year before. Critically, 94% of buyers rank vendors in preference order before contacting them, of which the pre-contact favorite wins 80% of the time.
Those numbers deserve more than a moment’s reflection. If the shortlist is formed before you can see the buyer, and if the pre-contact favorite wins four out of five times, then the battle for pipeline is being fought in channels your attribution model cannot access.
How are B2B buying decisions really made?
The Yes Advantage is Transmission’s research into the hidden forces that shape B2B purchase decisions — including the shortlist dynamics, peer influence, and brand signals that determine who wins before the buying cycle officially begins.
Taking action: tracking the modern B2B buyer journey
Accepting this doesn’t mean accepting defeat. It means reframing dark intent as a known variable that shapes how you interpret the data you do have. Understanding when and where the dark portions of your journey exist enables two distinct actions.
The first is a target for qualitative research. If you know buyers are forming shortlists in peer communities, LinkedIn DMs, and AI tools before they become visible, you can design research programs – customer interviews, focus groups, win/loss analyses – specifically to illuminate those stages.
What communities are your buyers active in? What questions are they asking AI tools? What peer recommendations carry weight? These are questions with answers, even if said answers require qualitative analysis over analytics dashboards.
“In an era where genAI shapes vendor shortlists before buyers ever raise their hand, brand strategy isn’t a ‘nice to have’ – it’s the thing that determines whether you’re considered at all. If buyers are forming views about your brand in private channels, AI outputs, and peer conversations you can’t track, then your brand needs to be doing consistent, compelling work in every channel you can influence. The brands that win the Day One shortlist discussion are the ones that have been building credibility long before the buying cycle officially begins.”
The second is better contextual interpretation of the data you can see. If you know that a significant portion of your ‘Direct’ traffic is actually AI-referred, and that the buyer who just filled in a contact form has probably been researching you for six weeks in channels you can’t see, you make better decisions about nurture sequences, sales follow-up timing, and content investment.
Adding a simple “How did you hear about us?” open-text field to your conversion forms isn’t a sophisticated analytics play. But it will tell you things about dark intent that no tracking implementation can.
Moving towards a future-proofed measurement framework
The good news, and there genuinely is good news, is that measurement maturity is a solvable problem. It doesn’t require perfect data. It requires a more honest, more sophisticated relationship with the data you already have.
BCG and Google’s research on digital marketing maturity established a benchmark that remains highly relevant: organizations with mature cross-channel measurement significantly outperform peers on revenue growth and marketing efficiency.
The path from siloed channel metrics to full-funnel optimization is well documented. What the 2025–2026 environment adds is a new dimension – AI visibility monitoring – that didn’t exist when that framework was developed.
Measurement maturity in the current landscape moves through recognizable stages. Most B2B marketing teams are still working at the level of channel-specific metrics (organic traffic, paid conversions, email open rates) without a coherent model for how those channels interact.
The next stage is cross-channel attribution that acknowledges the limitations of last-click models and attempts to weight touchpoints more accurately. Beyond that sits predictive modelling, where historical patterns are used to forecast pipeline contribution from channels that resist direct attribution.
And at the most mature end is full-funnel optimization that incorporates brand measurement, qualitative insight, and AI visibility alongside performance data.
Key components of a mature B2B measurement framework in 2026
- A first-party data strategy that doesn’t depend on third-party cookies, capturing declared intent through form fields, preference centers, and direct buyer conversations
- Qualitative research programs designed specifically to illuminate dark journey segments – the touchpoints that precede and follow every gap in your analytics
- Attribution models that present ranges and confidence levels rather than false precision, because leadership making decisions from overconfident numbers makes worse decisions than leadership that understands the uncertainty
- Brand measurement that captures influence beyond direct response, including share of voice in AI outputs, analyst citations, and community presence
- AI visibility monitoring as a genuinely new measurement dimension, tracking where and how your brand is represented in the LLM responses your buyers are reading
Building a brand influence architecture for your B2B buyer journey
The closing message for CMOs building this framework is the one the data keeps returning to. Buyers form their shortlists in channels you cannot track. The pre-contact favorite wins. The battle for pipeline is fought in dark social, AI tools, peer communities, and internal conversations that leave no measurable trace.
Winning that battle requires building what amounts to an influence architecture – sustained presence and genuine authority in the places where decisions actually form – and accepting that marketing’s true contribution to revenue will always be larger than any attribution model can prove.
Mature measurement doesn’t close that gap entirely. But it gets you close enough to make better decisions, tell a more credible story to the board, and invest with confidence in the channels that matter – including the ones you can’t see.
Ready to improve how you track & influence the B2B buyer journey
We build the measurement frameworks, brand strategies, and digital optimization programs that keep you visible, and influential, at every stage of the buyer journey – including the stages you can’t directly track.