The Buyer Left the Funnel.

Executive Summary

Revenue leaders are fighting what looks like a dozen separate fires: pipeline is harder to source, customer acquisition costs keep climbing, budgets shrink while the software stack sprawls, and teams quietly blame one another for the gap between plan and result. Treated separately, each gets its own task force and its own tool. Treated together, they resolve into a single cause.

The buyer left the funnel. The funnel was a seller’s model of a buyer’s journey: orderly stages we could meter and manage. Buyers now run their own process: they research without us, shortlist without us, and increasingly let a machine form the first impression of our category and our company. Most of the decision is made before a salesperson is ever contacted.

This report is in two parts. Part I diagnoses the structural shift across five chapters. Part II is the operator’s response in five moves, followed by a phased agenda for sequencing the work.

The Mismatch

Why the motion that built the category now works against it, in five chapters of evidence.

From funnel to a buyer-controlled journey.

The defining change is one of control. A buying group of six to ten people now self-directs research across communities, peers, search, and AI assistants, assembling a point of view long before they raise a hand. By the time sales is invited in, the shortlist usually exists and the criteria are largely set.

Two-thirds of buyers say they would prefer to complete a purchase with no sales rep at all, and a similar share now expect a fully digital, self-serve path. The seller’s job moves upstream: you have to be present and credible in the journey, not waiting at the bottom of a funnel that buyers no longer descend in order.

A machine forms the first impression.

A new layer now sits between your company and the buyer: the generative answer engine. When a buyer asks an assistant to frame a problem or name the vendors that solve it, the model’s synthesis, not your homepage, is the first impression. Adoption moved from near-zero to the high eighties inside a single year, and generative AI is now cited as a leading research source.

The risk is specific and measurable: what the machines say about your brand is often wrong: outdated positioning, missing products, competitors’ claims attributed to you. You cannot opt out of this layer; you can only choose whether to be represented accurately in it.

The math stopped working.

As the old motion loses efficiency, its cost rises. Many software companies now spend on the order of two dollars to acquire a dollar of new ARR, a ratio that only survives in a cheap-capital world. That world is gone, and the finance function has noticed.

Marketing’s share of revenue has compressed: martech budgets fell from roughly 9.5% to 7.7% of revenue, while broader IT and business-services spend slid from about 9% to 5.8%. Most marketing leaders now report being underfunded against their mandate. Efficiency is no longer a virtue; it is the condition of survival.

More tools, less leverage.

The reflex response to a broken motion is to buy software. The result is a martech landscape of roughly fourteen thousand tools, of which the average organization deploys about ten and genuinely relies on four or five. The rest is cost, integration debt, and cognitive load.

The evidence points the other way: teams that rationalize their stack, cutting to what they actually use, see materially more pipeline per head. The discipline is subtraction, even as the category’s total spend is projected to pass $215B by 2027.

The perception gap.

Underneath the numbers is a human problem. Most front-line teams see clear misalignment between marketing, sales, and product. Most executives believe their teams are aligned. Both cannot be right, and the gap between them is where strategy quietly fails to execute.

Only a small minority of organizations report genuinely strong alignment. The fault line isn’t a culture issue to be smoothed over with an offsite; it is a structural symptom of teams optimizing for different, funnel-era metrics that no longer add up to a single journey.

The Playbook

Five moves operators can sequence, from creating demand to owning a category and the way it is measured.

Most of the market isn’t buying today.

At any moment only a small fraction of your market is in an active buying cycle; the rest are out-of-market and will be for months or years. A go-to-market system tuned only to capture in-market demand competes for that sliver at ever-rising prices. The durable move is to create demand: building memory and preference now, so you are the default when buyers enter the market later.

The classic brand-to-activation balance, roughly 60:40 in favor of long-term brand building, exists precisely because the 95% who aren’t ready today decide who they trust tomorrow.

Write for the engine that reads you.

If a machine forms the first impression, then the machine is an audience, and it can be optimized for. Generative engine optimization is the discipline of being accurately understood, retrieved, and cited by answer engines: clear claims in real text, structured evidence, consistent naming, and sources a model can trust.

This page is built to its own advice: semantic headings, selectable text for every statistic, descriptive anchors, and a visible source-and-method block, so it is citable by both people and engines.

The winner defines the question.

Markets reward the company that frames the problem, not merely the one with the best feature list. The Category King, the brand that names and defines a category, captures the lion’s share of its economic value. Category design is therefore not a branding exercise; it is the highest-leverage demand-creation move available.

Designing a category means choosing the problem you want to own, giving it language, and making that language the default frame buyers, and machines, reach for.

Three moves to fit the new journey.

The fault line in Chapter 5 is structural, so the fix is structural. Three moves re-point the organization at the journey buyers actually run:

Measure the journey, not the funnel.

You manage what you measure, and funnel-era metrics quietly pull the whole organization back toward the funnel. The mandate has to change, and so do the numbers that define success.

These measures are harder to game and slower to move, which is exactly why they reflect the durable health of a go-to-market system built for a buyer-controlled, machine-mediated journey.

Where to start, in order.

Three phases, sequenced so each earns the right to the next. Don’t skip Phase 1.