The marketing funnel is nice, but it misses a very crucial aspect: the why
The funnel is a model of the business, not of the customer's mind. Bob Moesta's demand-side thinking shows what is happening underneath each stage, and why that changes demos, content, search and sales.
5 October 2026 · 14 min read
Marketing is hard, which is why the best way to make sense of things is to turn complex human behaviour into diagrams. Few attempts at this are more successful than the funnel. Awareness sits at the top, followed by interest, consideration and intent, until eventually the fortunate few emerge from the bottom as customers. It is useful, memorable and remarkably durable. It is also one of those models that becomes less useful the more literally we take it.
Why is the Marketing Funnel Fundamentally Wrecked?
The funnel describes the customer from the company's perspective. It tells you how many people entered, how many progressed, where they dropped off and how efficiently the business converted one stage into another. What it doesn’t tell you particularly well is what was happening in the customer's world while all of this was taking place.
Bob Moesta and His JTBD Answer
A person in the "consideration" stage might be learning that a problem exists, comparing different alternatives, trying to convince their boss, worrying about implementation, or simply satisfying their curiosity. The funnel gives all of those people roughly the same label.
That distinction became much clearer to me through the work of Bob Moesta, particularly his work on Jobs to Be Done and what he calls Demand-Side Sales. Moesta's starting point is relatively simple: customers are trying to make progress, not trying to move through your funnel.
This may sound a lot like a semantic postulation, but when you start applying it to an actual business, it starts making sense. It changes how you think about demos, websites, content, search, sales calls, onboarding, product design and even what constitutes a "qualified" customer.
Funnel Is To Business, Not Customer
There is nothing inherently wrong with measuring awareness, leads, opportunities, purchases or retention. A business needs to know these things. If 10k people visit a website and only 20 become customers, there is a commercial problem worth investigating. If a SaaS company generates plenty of trials but almost nobody becomes a paying customer, that matters. The problem begins when we assume that the stages themselves explain customer behaviour.
Take the word "consideration." It sounds precise because it’s a familiar marketing term, but it tells us very little. One person might be considering whether they have a problem at all. Another may have spent six weeks researching solutions. A third might have already chosen a category and be deciding between two vendors. A fourth may know exactly what they want but be waiting for procurement to approve the purchase. All four can appear somewhere in the same funnel stage.
Google's research into what it calls the "messy middle" makes a similar observation from a different direction. Its research found that the space between a purchase trigger and a final decision is not linear. Consumers move between exploration and evaluation, often repeatedly, while consulting search engines, reviews, social platforms, aggregators and other sources. Google describes this as a complex space in which customers are effectively won and lost.
The implication isn’t that funnels should be abolished. It’s that we should be careful about confusing the company's reporting structure with the customer's mental and behavioural process. Moesta's work gives us another way of looking at that process.
Customers Make Progress. They Don't Move Down Funnels
One of the most useful ideas in Moesta's work is the buying timeline. Rather than assuming that every prospect should be pushed through the same sales sequence, he describes a progression that includes first thought, passive looking, active looking, deciding, first use and ongoing use. The important point is that the customer has their own timeline, and the company's sales process works best when it accommodates that timeline rather than forcing everybody through an identical experience.
That changes the question you ask.
Instead of asking, "How do we move this lead to the next stage?" you might ask, "What is this person trying to figure out right now?"
Perhaps they’ve only just noticed that something isn't working. They need language for the problem.
Perhaps they know the problem very well but have no idea what their options are. They need education and contrast.
Perhaps they have already explored the category and are comparing alternatives. They need help understanding trade-offs.
Perhaps they have made a decision in principle but are worried about implementation. They need reassurance, evidence and a credible path to adoption.
The company's funnel might describe all of these people as prospects. The customer is having very different experiences.
Moesta's four forces of progress provide another useful lens. There is the push of the current situation, the pull of the new solution, the anxiety associated with changing, and the habit or inertia of the present situation. In other words, people are not simply persuaded into buying because a product looks attractive. They are weighing the discomfort of staying where they are against the perceived benefits and risks of changing. This is an important corrective to a lot of modern marketing.
When conversion falls, our instinct is often to add more pull.
Better copy.
More features.
A stronger offer.
Another case study.
A bigger discount.
A more compelling CTA.
But perhaps the customer already understands the attraction of the product. Perhaps their problem is anxiety.
"Will this integrate with our existing system?"
"Will my team actually use it?"
"What happens if we migrate and something breaks?"
"Will I look stupid if I recommend this internally and it fails?"
Or perhaps the problem is habit.
"We have always done it this way."
"It is annoying, but it works."
"We don't have the time to change."
More product benefits will not necessarily solve either problem.
The Idea In Motion
Moesta's Autobooks example is one of the clearest illustrations of this idea.
Autobooks had a problem many businesses can relate with. They had a demo, and the demo wasn't working particularly well. The instinct was to keep reworking the demo. Moesta asked a question: whose timeline was the demo designed around?
The company had one demo, but the customers didn’t have one buying timeline. So the team began distinguishing between different points in the customer's process. The first demo was about learning, the second was about possibilities, and the third was about trade-offs. Instead of forcing everyone through one standard presentation, the interaction changed according to where the buyer actually was.
The results Moesta presented were striking. The company cut the sales cycle from lead to customer to roughly half its previous length, while lead conversion roughly doubled in the earlier version of the case. In later presentations, he reported even stronger results, including a threefold increase in lead conversion and a halving of time to close.
The important lesson isn't "three demos are better than one." That would be missing the point entirely. The lesson is that the business stopped asking, "How do we get everyone through our demo?" and started asking, "Where is this customer in their buying process?"
The Fault In Tech Companies’ Stars
Tech companies are particularly vulnerable to the funnel problem because software businesses tend to have beautifully measurable stages. Visitor. Lead. MQL. SQL. Demo. Trial. Opportunity. Closed-won. Activation. Expansion.
But consider three people who request a demo of the same B2B software.
The first has just realised that their existing process is breaking down. They are trying to understand the problem and whether software is even the answer.
The second has already decided that they need a solution. They have looked at four competitors and want to understand what is possible.
The third has selected a shortlist and needs to determine which product is safest to introduce into the organisation.
All three have requested a demo. A conventional sales process might give them the same 45-minute presentation. A demand-side approach would begin somewhere else: "Why are you looking now?"
That question can reveal the struggling moment, the desired progress and the forces surrounding the decision. The resulting conversation can then be much more useful.
This applies beyond sales. A free trial is not automatically a "trial stage" in some universal customer journey. One person might be experimenting, another is trying to complete a specific workflow, another is comparing you with a competitor, while another signed up because a colleague sent them a link and they have no intention of becoming a customer.
If activation is poor, the question should therefore not simply be, "How do we increase activation?" It should be, "What were people trying to accomplish when they signed up, and where does the product fail to help them accomplish it?"
Examples In Motion: Marketing
Agencies have their own version of the funnel.
Awareness campaigns generate traffic, lead magnets generate leads, nurturing produces opportunities, sales calls produce clients, retainers produce recurring revenue.
But imagine an auto shop owner approaching a marketing agency.
"I've heard SEO is important, but I'm not sure what I actually need."
That person is not asking for an SEO strategy yet. They are trying to understand a problem.
Another says:
"We know competitors are taking searches from us. We've tried content and Google Ads, but we don't know what to fix."
That's a different job. They are looking for a diagnosis and possible paths forward.
Another says:
"I've talked to three agencies. Tell me why I should work with you."
That person isn't looking for an introduction to SEO. They are trying to resolve a choice.
The agency's CRM may classify all three as leads, but the customer's world does not.
This is also where I think the demand-side view gives SEO a more useful foundation. Search behaviour can often provide clues about where somebody is in their attempt to make progress. "What is SEO?" represents a different information need from "SEO agency for auto repair shops," which is different again from "SEO agency pricing," "Agency A vs Agency B," or a branded search for a company that someone has already heard about.
These aren't merely different keywords to optimise. They are different questions arising at different points in a decision.
The marketer's job, then, is not simply to "cover the funnel with content." It is to understand the questions that emerge as customers move through their situation and make the information, evidence and choices available when those questions arise.
That is a subtle distinction, but it changes how you build a content strategy.
In D2C
Ecommerce makes this particularly visible because the distance between consideration and transaction can sometimes be measured in minutes.
Imagine somebody buying running shoes. The funnel might tell us they became aware of the brand, considered several products, visited a product page, added something to their cart and eventually purchased.
But what actually happened?
Perhaps they have recently started running and their current shoes hurt.
Perhaps they have a race in six weeks.
Perhaps they have read that they need stability shoes but don't know what that means.
Perhaps they have narrowed their choice down to two models and are terrified of spending $180 on the wrong one.
The job is not simply "buy running shoes."
It might be: "Help me choose something I can run comfortably in without making an expensive mistake."
That distinction changes the website.
Reviews become more than social proof. They answer the anxiety of, "Will this work for someone like me?"
Sizing information answers, "Will this fit?"
Comparison tables answer, "How do I choose between these two?"
Returns information answers, "What happens if I get this wrong?"
Product demonstrations answer different questions depending on the customer's uncertainty.
In other words, good ecommerce does not merely push the customer toward checkout. It removes the things standing between the customer and a confident decision.
In enterprise, there may not be one customer
Enterprise software introduces another complication: "the customer" may actually be several people with several different jobs.
The person who uses the product wants one thing.
The person managing the department wants another.
The CFO wants to understand the economics.
IT wants to understand security and integration.
Procurement wants predictable terms and a manageable vendor relationship.
The executive sponsor wants to know whether the purchase advances an organisational priority.
A conventional account funnel can represent these people as one opportunity moving toward a close. Demand-side thinking encourages you to ask what progress each participant needs to make.
That has obvious implications for sales enablement. Instead of creating "consideration-stage content," you might create an implementation guide for IT, an economic case for finance, a workflow demonstration for the end user and a strategic argument for the executive sponsor.
The account is moving toward a purchase, but the people inside the account are solving different problems.
This is one reason B2B purchasing can feel irrational when viewed only through a funnel. The organisation may desperately need a solution while still failing to buy it because one participant has unresolved anxiety, another prefers the existing system, and another cannot justify the investment internally.
The deal isn't necessarily "stuck in consideration." The people involved are struggling to make progress.
And in local business, the timeline can be brutally short
The same thinking works at the other end of the market.
Consider local search.
Someone searches "why does my tooth hurt" and someone else searches "dentist near me." They may both eventually become dental patients, but they are not necessarily at the same point in their decision.
One is trying to understand what is happening, the other may already have decided that they need a dentist and simply wants to find one.
Likewise, "brake squeaking" and "brake repair near me" may belong to the same service category while representing very different moments in a customer's timeline.
The distinction matters because the customer's information needs change.
A person still diagnosing a problem needs understanding.
A person choosing a provider needs confidence.
A person comparing prices needs a different kind of information.
A person who has already decided to call needs friction removed.
The funnel can tell a local business that search traffic became a phone call. The demand-side view encourages us to ask what happened before the phone call.
So what should marketers actually do differently?
For me, the practical implication is not to throw away the funnel. It is to put another layer underneath it.
When I look at a customer or prospect, I want to understand at least five things.
First, what changed? Something usually creates the space for a new solution. It could be frustration, growth, an event, a failure, a new responsibility or a new possibility.
Second, what are they trying to accomplish? Not what product category are they researching, and not what feature are they clicking, but what progress are they actually trying to make?
Third, what alternatives have they considered? The alternative may be a competitor, but it may also be doing nothing, continuing with spreadsheets, hiring someone, asking a friend, repairing the existing thing or tolerating the problem.
Fourth, what makes changing difficult? This is where anxiety and habit become important. The customer may understand your value proposition perfectly well and still not buy because changing feels risky or inconvenient.
Finally, what do they need next? What information, proof, demonstration, comparison, reassurance or action would help them make the next piece of progress?
These questions produce a different kind of marketing.
Instead of asking, "What should we say to this audience?" we can ask, "What are they trying to figure out?"
Instead of asking, "How do we improve the demo?" we can ask, "What does this particular buyer need from the demo?"
Instead of asking, "How do we get more people through checkout?" we can ask, "What uncertainty is preventing people from completing a purchase?"
Instead of asking, "Why aren't our trials converting?" we can ask, "What were these people trying to accomplish when they signed up?"
The latter questions tend to produce better work because they start with the customer's reality.
The funnel still matters
I don't think the answer is to declare the funnel obsolete. That would simply replace one oversimplification with another.
A business still needs to understand acquisition, conversion, revenue, retention and economics. Leadership still needs dashboards. Marketing teams still need to know where demand is being generated and where it is being lost.
The funnel is very good at telling us where the commercial system has a problem.
What it is less good at is telling us why the customer is behaving that way.
That is where customer research, Jobs to Be Done and Moesta's demand-side thinking become useful.
If trial-to-paid conversion falls, the funnel gives you the location of the problem.
Customer interviews might reveal that users never encountered the feature that demonstrates value.
Or that they were merely experimenting.
Or that implementation anxiety is killing the decision.
Or that the product solves a problem that customers do not consider sufficiently important to change their behaviour for.
The funnel tells you where to look.
The customer's story tells you what to look for.
That distinction is easy to miss because marketers are often trained to think about movement. We are taught to move people from awareness to consideration, from consideration to intent, from intent to purchase. Moesta's work asks us to pause before doing that and understand the movement already happening in the customer's life.
That is a considerably more interesting way to think about marketing.
Customers aren't trying to complete our funnels.
They are trying to solve problems, reduce uncertainty, make decisions, avoid mistakes, change something that has stopped working and get on with whatever they actually care about.
The funnel remains useful. It tells the business how its commercial system is performing.
But it should remain what it is: a model of the business, not a model of the customer's mind.
And perhaps the better question for marketers is not, "How do we move this person to the next stage?"
It is:
"What progress are they trying to make, and what is standing in their way?"
That is where the marketing work begins.
Where this shows up in the work
Keep reading
- How to actually pull off Jobs To Be Done (without losing your mind or building features nobody wants)A practical field guide to JTBD: how to run switching interviews, map the four forces, avoid the classic traps, measure real progress, and turn it all into product, marketing and sales decisions.
- Personas describe people. Jobs explain decisions.JTBD is a lens, not an identity. Used well it explains why someone switched; used badly it becomes another artefact nobody opens.
- Positioning is not copywritingIf the words outrun the product, you have bought yourself an expensive disappointment. Positioning should be the shortest accurate description of real value.