The 70 percent problem: why your pipeline reporting misses most B2B buyers

The 70 percent problem is the gap between how B2B buying actually happens and what your pipeline reporting can see. Most of a purchase decision, the research, the shortlisting, the internal debate, happens before anyone fills in a form. By the time a buyer becomes a lead in your CRM, the bulk of the work is done where your reporting cannot reach it. Your dashboard sees the last act and misses the play.

What is the 70 percent problem in B2B?

It is the reporting blind spot at the centre of most B2B marketing. Buyers do the majority of their evaluation alone and in private, then surface late, if at all, as a form fill or a demo request. The number itself is shorthand for a well documented pattern. Gartner research found that B2B buyers spend only 17 percent of their total buying time meeting with potential suppliers, and when several suppliers are in the running, as little as 5 to 6 percent with any one sales rep (Gartner, B2B Buying Journey research). Around four fifths of the decision runs without you in the room.

So the pipeline you report on is the visible tip. A buyer reads three of your articles, asks two peers on a private Slack, listens to a podcast you were mentioned on, then shortlists you without a single tracked click. When they finally appear, the CRM logs them as a cold inbound lead. The reporting is not wrong about what it counted. It is wrong about what it implies, because it counts only the buyers who chose to identify themselves.

Why do most B2B leads never convert to pipeline?

Because the lead is rarely the buyer. A form fill is usually one person on a committee of several, and often the wrong one for the stage. Three patterns explain most of the leakage.

First, timing. The Ehrenberg-Bass Institute, in work for the LinkedIn B2B Institute, found that only about 5 percent of business buyers are in market at any given time, with roughly 95 percent not currently buying (Professor John Dawes, Ehrenberg-Bass Institute, 2021). A lead captured this quarter may genuinely buy in eighteen months. Score it as dead by Friday and you have thrown away a real future customer because the report had no column for patience.

Second, the unit of measurement. Most systems track individuals. B2B purchases are made by groups. One contact downloads a guide while five colleagues who never touched your site quietly decide the outcome around them.

Third, the channel mix. The activity that does the persuading often leaves no trace your analytics can read. That untracked space has a name.

What is the dark funnel, and why does it matter?

The dark funnel is the part of the buying journey standard analytics cannot see. Peer recommendations, private messaging groups, review sites, podcasts, communities, search in incognito, content forwarded person to person: all of it shapes the shortlist and none of it shows up as a clean source in your reporting.

This matters because the dark funnel is where trust gets built, and trust is what decides B2B deals. When a buyer finally lands on your site and converts, last click attribution hands the credit to whatever they typed into Google. The podcast, the recommendation and the six months of reading that actually moved them get nothing. Optimise on that flawed signal and you defund the work that wins deals while pouring budget into the channel that merely closes the tab. We wrote about this distortion at the channel level in our guide to why performance marketing should optimise for pipeline, not clicks, and it is the same error showing up one layer higher in the report.

What is a buying committee, and why does it break pipeline reporting?

A buying committee is the group inside a company who shape and approve a purchase, commonly six to ten people across functions: the user, their manager, finance, IT, procurement, sometimes legal. Each has a different question and a different fear. Most pipeline reporting cannot hold this, because it was built to track a lead, not an account.

So the committee fractures across your funnel. One member is a marketing qualified lead. Two are anonymous sessions. The economic buyer never visits your site at all and forms a view entirely from what the others relay. Your report shows one lukewarm contact at an early stage. The reality is an account in active evaluation with most of its decision makers invisible to you. This is the MQL to pipeline gap that frustrates so many B2B teams: the leads convert poorly not because the leads are bad, but because a single lead was never a fair picture of a group decision.

How does AI change what you can measure?

AI does not switch a light on in the dark funnel. Nothing does, and any vendor promising full visibility is selling you a story. What good AI implementation does is read the faint signals you already collect and join them into a picture closer to how buying really works.

Three shifts matter. AI moves the unit of measurement from the lead to the account, so six fragmented contacts read as one committee in motion rather than six unrelated names. It blends intent data, third party account signals and your own first party behaviour into a single score, so a quiet account showing a spike in research surfaces before anyone fills in a form. And it reads pipeline attribution as a pattern across many touches instead of crowning the last click, which gives the dark funnel partial credit for the influence it earned.

This depends on one thing being right first: your data has to be joined up. Account signals living in one tool and web behaviour in another cannot be scored together. That groundwork sits inside the wider practical 2026 framework for implementing AI in a B2B business, and it is the same plumbing that makes AI lead scoring and ABM intelligence work in practice rather than in theory. Modern tooling helps here. Models such as Claude, connected to your systems through the Model Context Protocol, can pull account history, support tickets and product usage into one view a salesperson can act on, the worked example we cover in the lead scoring post.

There is a tidy reason to fix this now. Buyers have moved their early research into AI search, and they are doing it in the dark funnel by default. Around 89 percent of B2B buyers use AI search during the buying process, and AI Overviews trigger on roughly 48 percent of tracked queries (Digital Agency Network, generative engine optimisation statistics, 2026). The first time many buyers meet you in 2026 is in an answer from ChatGPT or Google, weeks before they would ever count as a lead. If your reporting only respects form fills, that entire first encounter is invisible to the people deciding your budget.

What should B2B teams do about the 70 percent problem?

Start by being honest about what your pipeline report is: a record of the buyers who put their hand up, not a measure of demand. Then close the gap in three moves.

Measure at the account level. Roll individual contacts up to the company so a committee reads as one opportunity rather than six scattered names. This alone changes which deals look alive.

Give the dark funnel credit it can earn. Add self reported attribution, a simple how did you hear about us field, and treat brand search and direct traffic as downstream of influence rather than as free wins. The point is to stop punishing the channels you cannot perfectly track.

Build on joined up data. Account scoring and intent only work when your tools share one view of the customer. Getting that foundation right is the unglamorous part, and the part that decides whether everything above is real or a slide.

The cleanups compound. When HMS Networks worked with us on exactly this kind of pipeline focused programme, joining the data and scoring accounts rather than chasing clicks, the result was 3,000 qualified sales opportunities in 90 days and a 73 percent cut in cost per sales lead. What worked was simple: they measured the thing that mattered and built on data that agreed with itself.

If you want help closing the gap, it sits inside the B2B lead generation programme and the AI implementation services we run for clients, and it pairs naturally with the AI tool stack audit when your reporting is fragmented because your tools are. The 70 percent problem will never disappear entirely. Most of B2B buying will always happen out of sight. The win is to stop pretending the visible 30 percent is the whole market, and to build reporting honest enough to act on the rest.

Frequently asked questions

What is the 70 percent problem in B2B?
The 70 percent problem is the gap between how B2B buying actually happens and what your pipeline reporting can see. Most of a purchase decision, the research, the shortlisting and the internal debate, happens before anyone fills in a form. By the time a buyer becomes a lead in your CRM, the majority of the work is done where your reporting cannot reach it, so your pipeline shows a fraction of the real demand.

What is the dark funnel?
The dark funnel is the part of the B2B buying journey you cannot track with standard analytics. It includes peer conversations, Slack and WhatsApp groups, podcasts, review sites, search done in incognito, and content shared person to person. Buyers move through it for weeks or months before they ever click a tracked link, so it shapes the decision long before your CRM records a single touch.

Why do most B2B leads never convert to pipeline?
Most B2B leads never convert because the lead is not the buyer. A form fill is usually one person on a buying committee of six to ten, often a researcher rather than a decision maker, and frequently captured long before or long after the moment that mattered. Pipeline reporting that treats each lead as a single person at a fixed stage misreads a group decision that has been running quietly in the background.

What is a buying committee, and why does it break pipeline reporting?
A buying committee is the group of people inside a company who shape and sign off a B2B purchase, typically six to ten individuals across different functions. It breaks pipeline reporting because most systems track individual leads, not accounts. One contact enters your funnel while the rest of the committee stays invisible, so the report shows one lukewarm lead when an account is in fact deep in an active evaluation.

Can AI fix B2B pipeline attribution?
AI cannot make the dark funnel fully visible, but it narrows the gap. By joining intent data, account signals and first party behaviour across your tools, AI can score accounts rather than isolated leads and flag when a committee is heating up before a form is filled. It moves reporting from counting clicks towards reading buying intent, which is a closer match to how B2B purchases actually happen.