Discover why modern buyers follow nonlinear journeys and how AI search, brand trust, and connected marketing reshape the traditional funnel.
For marketing leaders whose budgets, teams, and dashboards are still organized around a funnel their buyers never walked.
For more than a century, marketing has been drawn as a funnel: wide at the top, narrow at the bottom, with awareness leading to consideration, then intent, then a sale. You have probably drawn it yourself, on a whiteboard, in a deck, or in a budget. It feels true.
It is a useful picture of volume. It has never been an accurate picture of how one person buys. Real buyers do not slide down a funnel. They get curious, lose interest, and come back a week later. They decide, then change their minds. They read the fine print before they have heard of you, then circle back to the basics. They skip steps, repeat steps, and sometimes start at the bottom.
For a long time, the funnel was close enough to be useful. AI assistants are making the gap too wide to ignore. This post explains where the funnel came from, why companies held on to it, what replaces it, how to measure and defend the change in front of a board, and why brand is the one asset that works at every turn of the journey that actually exists.
From funnel to loop
The loop changes the work, not just the diagram.
-
Funnel
Sequence
Fill the top, push people down.
Loop
Presence
Be useful when buyers explore and when they evaluate.
-
Funnel
Stage conversion
Count people moving from step to step.
Loop
Presence and trust
Share of search, shortlist presence and how AI answers describe you.
-
Funnel
Stage teams
Top team, bottom team, split budgets.
Loop
The whole journey
Shared capabilities that serve exploring and evaluating.
would keep using a brand that starts selling to people different from them, among those who trust brands and find them relevant, versus those who do neither. Trust holds buyers as a brand changes.
Keep the word, change the plan
Use the funnel to talk about volume. Use the loop to plan how you reach buyers.
Source: Edelman Trust Barometer Special Report, June 2026, survey in 15 countries.
The funnel described a process, not a person
It began as a sales aid for picturing many prospects, not a study of how individuals decide.
The idea traces back to the 1890s, when the advertising pioneer E. St. Elmo Lewis described a sequence for persuading customers that later became known as AIDA: attention, interest, desire, action. The funnel drawing came later, as a way to picture how a large pool of prospects narrows to a smaller pool of buyers. It described flow across many people in total. It was a model of volume, not a model of a mind.
The mistake came afterward, gradually. Companies started treating the picture as the person. We acted as if each buyer really did move in order from awareness to purchase. We built teams around that sequence, split budgets by its stages, and mapped every piece of content to a step on the line. The buyer was looping the whole time. We just kept drawing them descending.
So the AI era did not break the funnel. It revealed that the funnel was never an accurate map of individual buying in the first place.
Why companies kept the funnel long after buyers left it
The funnel survived because it matched how companies count, staff, and pay.
The funnel lasted because it was useful to the company, even when it did not describe the customer. A funnel turns messy behavior into stages a team can count, and once something can be counted it can be given a target, a budget line, and an owner. Most CRM systems (customer relationship management software, the tool a sales team uses to track deals) are built around pipeline stages, so the software itself asks every deal to sit in one stage at a time and move forward.
The handoffs between teams reinforce the same picture. Many companies still define a marketing qualified lead (MQL, a prospect marketing judges ready for sales) and a sales qualified lead (SQL, a prospect sales has accepted), and they set targets and pay on how many of each cross the line. That structure assumes a buyer moves once, in one direction, from one team to the next. A buyer who reads three comparison articles, goes quiet for two months, asks an AI assistant about your pricing, and then shows up ready to talk does not fit a single stage, so the system either loses that history or files it under whichever touch happened last.
Attribution models (the rules that assign credit for a sale to earlier marketing touches) complete the picture. Last-click and stage-based models reward whatever happened closest to the sale, which tends to move budget toward the bottom of the funnel year after year, because that is where credit is easiest to see. The result is a company that, in essence, keeps funding the part of the journey it can measure while underfunding the part that built the buyer's preference. The people who built these systems made reasonable choices for their time, but the systems describe the company's workflow, and the buyer was never part of the design.
The loop is what was always there
Buyers move back and forth between exploring options and narrowing them down.
In 2020, Google published research on how people actually make purchase decisions, conducted with the behavioral science consultancy The Behavioural Architects. It described the space between the first trigger of interest and the final purchase as "the messy middle." In that space, people move between two modes: exploring, which expands their options, and evaluating, which narrows them. They cycle between the two, sometimes many times, until they feel confident enough to buy.
That is a loop, not a funnel. There is no smooth slide through stages, only a buyer moving between two modes until they are sure.
The data on business buying points the same way. McKinsey's 2024 B2B Pulse Survey of nearly 4,000 decision-makers found that B2B buyers now use an average of ten channels in their buying journey, up from five in 2016, and that 42% used more than eleven different touchpoints. Twice as many channels means twice as many places where a buyer can stop, compare, and circle back.
The buyer is also rarely one person. Forrester's The State of Business Buying, 2026, published in January 2026, found that the typical B2B purchase now involves 13 internal stakeholders and nine external influencers, and that the number rises for complex or strategic purchases. Each of those people runs a smaller loop of their own, and the group's decision comes from all of those loops overlapping, which is why a single stage label on an account in your CRM rarely tells you where the decision actually stands.
Gartner reported in March 2026 that 67% of B2B buyers prefer a rep-free buying experience, based on a survey of 646 buyers conducted in August and September 2025, and that 45% of those buyers had used AI during a recent purchase. Buyers are doing more of the looping on their own, across more channels and with more colleagues, before they ever talk to a company.
Why AI widens the gap
An assistant lets a buyer collapse, reverse, or repeat the whole journey in one sitting.
For a century, the funnel was imperfect but workable. AI assistants are making it unworkable. A buyer can now fold awareness, consideration, and intent into one conversation. They can start at the bottom, asking for a specific recommendation, then work backward to learn the basics. They can loop through an assistant's answers several times in one sitting, comparing, doubting, and asking again.
Consider a typical buyer. She asks an assistant for the three best tools for her problem and gets names she has never heard of. She asks why one is better than another. She asks about price. She leaves, thinks about it, and comes back the next day to ask the same question a different way. She never saw an ad, filled out a form, or touched a stage on anyone's chart. She looped, quietly, until she was sure.
This behavior is now common rather than early. In Forrester's 2025 Buyers' Journey Survey, 94% of business buyers said they use AI in their buying process, up from 89% the year before, and twice as many buyers named generative AI or conversational search as their most meaningful source of information as named any other source, well ahead of vendor websites, product experts, and salespeople. For many purchases, the assistant is now the first place a buyer explores, before your website or your sales team has a chance to frame the problem.
The same research explains why AI produces more looping instead of a straight path. Buyers use AI to move quickly, but they do not fully trust what it tells them, so they check its answers with people. Gartner found in May 2026 that 69% of B2B buyers prefer to validate AI-generated insights with sales reps, and that 51% of buyers said they were more likely to run into misleading information from generative AI, compared with 49% who said the same about sales reps. A buyer explores with an assistant, evaluates with a colleague or a rep, returns to the assistant with sharper questions, and repeats the cycle until the doubts run out, which is the loop running faster than any stage-based plan was built to follow.
When the journey can collapse into a single chat, run backward, or spin in tight circles, the funnel does more than simplify reality. It misleads. A business organized around a misleading map will spend money in the wrong places, staff the wrong teams, and measure the wrong things.
The funnel told you where buyers should go. The loop shows you who you need to be when they get there, again and again.
Most of your future buyers are not shopping today
Buyers enter the loop on their own schedule, so the brands they already know usually get the first look.
In 2021, Professor John Dawes of the Ehrenberg-Bass Institute for Marketing Science described what is now called the 95:5 rule, in work published by LinkedIn's B2B Institute. The idea is simple arithmetic. If companies replace a type of product or service about once every five years, then only about 20% of potential buyers are in the market in a given year, and about 5% in a given quarter. The other 95% are out of the market for now, and no campaign will move them in, because buyers enter the market when their own situation changes, such as a contract ending, a new leader arriving, or a system failing.
Dawes presents the 95:5 split as a calculation, and the real figure depends on how often your category is bought, so it is worth running the math with your own average replacement cycle. The implication for the loop is direct. When one of those future buyers starts looking, the first thing they now do is ask an assistant, a colleague, or a search engine which options are worth considering, and the brands that come to mind or appear in those answers form the starting list. A plan that spends most of its effort on the 5% already in the market leaves that starting list to whichever competitors were building familiarity with everyone else in the meantime.
What the loop changes
Each shift moves budget, measurement, or ownership.
From sequence to presence. The funnel implies filling the top and pushing people down. The loop implies being useful and convincing in both modes, exploring and evaluating, at all times, because you rarely know where a buyer is. Someone who looks like a newcomer may have already evaluated you twice. In budget terms, this means funding the assets buyers return to in both modes, such as clear category explanations, comparison pages, pricing context, case studies, and third-party reviews, ahead of campaigns built for a single stage.
From stage metrics to presence and trust. The funnel counts people moving from one stage to the next. The loop asks different questions: whether you are present when buyers explore and when they evaluate, whether you are on the shortlist, whether you are named in AI answers, and whether your brand is searched for directly. With no clean stages to convert between, presence and trust become the measures that matter.
From stage-based teams to the whole journey. The funnel produces a top-of-funnel team, a bottom-of-funnel team, and budgets split by layer. The loop calls for organizing around the whole journey: building the content, proof, and experiences that serve exploring and evaluating, wherever buyers happen to be. In practice, that often means one shared owner for the content, proof, and product information that both marketing and sales draw on, so the story a buyer hears from your website, your sales team, and an AI answer stays the same. We cover the practical side of that, including how to connect channels and teams, in "A Guide to Connected Marketing."
How to measure presence and trust
Loop measures are less tidy than stage conversion, but several are well enough tested to report to leadership.
Share of search. Share of search is your brand's searches as a share of all brand searches in your category. Les Binet presented research at the IPA's EffWorks Global 2020 conference showing that share of search correlated with market share in the categories he studied (cars, home energy, and mobile phones), and that it could predict market share, in some cases up to a year ahead. This is a correlation, and it does not prove that searches cause sales, but it gives leadership an early read on brand demand that a stage report cannot.
Shortlist presence. In every win and loss interview, ask which vendors the buyer considered first and where that first list came from. Over a few quarters, the answers show how often you make the starting list and which sources, such as AI answers, peers, analysts, or review sites, are putting you there. We cover AI answer sampling and self-reported attribution in more detail in "The Invisible Journey."
Decision confidence. Gartner's March 2026 research found that buyers who reach high confidence in their decision are twice as likely to report a high-quality deal as buyers with low confidence. That makes the questions buyers keep looping back on a useful signal. If your sales team hears the same three questions on most calls, your content is most likely not answering them well enough in the places buyers explore.
As a practical rule, track these measures monthly for at least two full quarters before judging them, because loop measures move slowly and a single month of share of search data can mislead. This is a working guideline, not an official benchmark.
Brand is the asset that works at every turn
In a journey with no orderly path, a buyer who already trusts you leans your way wherever they are.
The funnel quietly downgraded brand. If you could catch people stage by stage, brand became a nice extra at the top while the real work happened lower down. The loop reverses that. When there are no tidy stages to catch buyers in, the one advantage that applies everywhere is a brand the buyer already knows and trusts.
Seth Godin has described a brand as the set of expectations, memories, stories, and relationships that together account for a customer's decision to choose one option over another. Read that with the loop in mind. None of those are stage tactics. They work across the whole journey, at any point, in any order.
The research supports the commercial case. Edelman's June 2026 brand trust report found that 71% of consumers would keep using a brand as it grows when they both trust it and find it relevant, compared with 36% when they feel neither. That is why brand demand, share of search, and a place on the buyer's day-one shortlist matter so much. They are the assets that work in a loop.
Keep the funnel for volume and present the loop to your board
The funnel is still a fair picture of totals. The work is to stop using it as a map of people.
The funnel does not need to vanish from your vocabulary. As a picture of volume, how a large pool narrows to a smaller one in total, it is fine, and it may help when talking with a board. The error is using it as a map of how one person decides, then building the company around that picture.
When you present the change, three points tend to carry weight with a board. First, the change corrects an outdated model, and the research behind it, from Google's 2020 work to Forrester's and Gartner's 2026 buyer studies, points in one direction. Second, the new measures are leading indicators, since share of search and shortlist presence tend to move before revenue does, which gives the board earlier warning than stage counts. Third, the change can be tested without restructuring the company, by running loop measures next to the existing funnel report for two quarters and comparing what each one predicted.
It also helps to name the cost of staying put. If most future buyers are out of the market today and will start their search with an assistant or a colleague, a budget weighted to the bottom stage spends most of its money on the few buyers already deciding, while competitors shape the starting list for everyone else. Keep the funnel slide for board conversations about volume, organize the work around the loop, and you can act on how buyers really behave while speaking a language your leadership already understands
The numbers behind the shift
More channels, fewer conversations
B2B buyers now use about ten channels to buy, double the number in 2016. 67% would rather not deal with a sales rep at all.
With that many touchpoints and fewer conversations, each channel has to tell the same story without a rep there to reconcile it.
| Finding | Figure | Source |
|---|---|---|
| Channels · McKinsey, nearly 4,000 B2B decision-makers in 13 countries | ||
| Average channels B2B buyers use in their buying journey, 2016 versus 2024 | 5 → 10 | McKinsey B2B Pulse Survey, 2024 |
| B2B buyers who used more than 11 different touchpoints | 42% | McKinsey B2B Pulse Survey, 2024 |
| Preference · Gartner, 645 B2B buyers | ||
| B2B buyers who prefer a rep-free buying experience | 67% | Gartner, March 2026 |
| Consumer context · Edelman, 15 countries | ||
| Willing to keep using a brand that starts selling to people different from them: those who trust brands and find them relevant, versus those who do neither | 71% vs 36% | Edelman, June 2026 |
McKinsey’s survey ran in 2024 and Gartner’s from August to September 2025. The Edelman figure describes consumers, not B2B buyers.
What this likely means over the next 6 to 18 months
Expect journeys to loop more, as AI assistants become part of everyday research.
If current trends continue, more buyers will research through AI assistants that let them jump between stages at will, and the number of channels and people in a typical journey is likely to keep rising. Forrester reports that procurement teams are now decision-makers in 53% of business buying cycles and engage from the start, and that more than 60% of buyers use a trial before they commit, rising to 78% for purchases of $10 million or more. Each of those steps adds another evaluation turn to the loop, which would make stage-based plans and metrics less reliable each year.
More of the looping is also likely to move inside tools you cannot see. Forrester found that 61% of business buyers already use private AI tools provided by their own organization, so a growing share of early research will happen in systems that draw on public, third-party information about your company rather than on anything you can target directly. The companies best positioned will most likely be those that measure presence and trust across the whole journey and invest in a brand, and a body of independent evidence, that buyers and their tools recognize wherever they are in the loop.
If you want help redrawing your journey map around how buyers actually decide, Camino5's Consumer Journey Strategy work starts there.
How to start today
Redraw your journey as a loop between exploring and evaluating, and share it with your marketing and sales teams, so both teams plan from the same picture of how buyers decide.
Review your content and sort it by mode: what helps a buyer explore, and what helps them evaluate. Fill the bigger gap first, because that is where buyers are most likely leaving to find answers elsewhere.
Add presence and trust measures, such as share of search, branded search, shortlist presence, and mentions in AI answers, to your monthly reporting next to your funnel metrics, so leadership sees both views together.
Ask in every win and loss interview which vendors the buyer considered first and where that list came from, so you learn which sources are putting you on the shortlist.
Look at how your budget is split between "top" and "bottom" teams, and identify one shared capability, such as comparison content or customer evidence, that serves the whole journey.
Keep the funnel slide for board conversations about volume, and bring the loop measures to your next board review as leading indicators.
The funnel will stay useful for counting totals, but the companies that plan around how people actually decide, moving back and forth between exploring and evaluating, are the ones most likely to be on the list when those buyers are ready.
Sources
McKinsey & Company, "Five fundamental truths: How B2B winners keep growing," 2024 B2B Pulse Survey
Think with Google and The Behavioural Architects, "Decoding Decisions: Making sense of the messy middle," 2020
Forrester, "B2B Buyers Make Zero-Click Number One," Buyers' Journey Survey, 2025
Gartner, "Gartner Sales Survey Finds 67% of B2B Buyers Prefer a Rep-Free Experience," March 9, 2026
IPA, "Binet presents fast, cheap, predictive Share of Search metric," EffWorks Global 2020
Edelman, 2026 Trust Barometer Special Report: Brand Growth in an Insular World, June 2026