Why is corporate video especially prone to the dark funnel?

Corporate video is especially prone to the dark funnel because it is usually commissioned for reach and impression, not capture. A brand film is watched, admired, and shared, but rarely wired to a tracked next step or a martech trigger. Without that plumbing, the view never resolves to a contact, so the spend cannot be traced to pipeline. Video also travels better than any other asset — into a feed, a deck, a forwarded file, a meeting-room screen — and every one of those destinations is somewhere your analytics does not reach.

How do you measure the dark funnel in your own video spend?

You measure it by testing three links: whether each video is commissioned against a defined funnel stage, whether your martech stack can attribute a view to a known contact and a tracked next step, and whether the brief tied the asset to a pipeline outcome. The Dark-Funnel Video Audit scores exactly these three links in two minutes. The rougher version costs nothing: count the deals your CRM shows a video touch on, then count the deals where a buyer mentioned video unprompted. The distance between those two numbers is your dark funnel.

How do you fix a dark-funnel leak?

You fix it by wiring one video end to end before scaling: define the funnel stage first, give the viewer a tracked next step, match the view to a known contact, and fire one automation on view. One instrumented asset teaches the stack more than ten untracked ones.

Can you eliminate the dark funnel completely?

No, and a plan that assumes you can will fail. Some of the most valuable video influence in B2B happens in places no instrumentation reaches — a film played in a boardroom, a clip forwarded inside a buying committee, a conference talk watched on someone else’s laptop. The realistic goal is to shrink the untraced share quarter over quarter and to report the remainder as a named quantity rather than pretending it is zero.

Is the dark funnel the same thing as dark social?

They overlap but are not the same. Dark social describes a distribution problem: sharing that happens over private channels where the referrer is stripped, so the traffic lands as direct. The dark funnel is broader — it covers every untracked influence between exposure and a known contact, including views that generate no click at all. For corporate video the distinction matters, because most video influence produces no session to misattribute in the first place.

Field guide

What is the dark funnel in corporate video?

By Paul Joseph · Updated 23 Aug 2026

The dark funnel is the untracked buyer activity between a viewed asset and a known contact. A prospect watches your brand film, forms an impression, and weeks later arrives through a direct visit, a search for your name, or a colleague’s referral — a path your analytics never connects back to the video. The influence was real. The attribution was invisible. Most corporate video spend lives entirely inside this gap.

If you would rather see the size of your own gap than read about the general case, the Dark-Funnel Video Audit scores it in two minutes and names the one link to wire first. It is free, and the rest of this page explains what the score means.

The path a view should take

Why most corporate video disappears into it

Corporate video is commissioned for reach, not capture. It is built to be watched and admired, then set loose on a channel with no tracked next step and no martech trigger behind it. That is a problem of plumbing, not craft. The 95-5 rule — from Professor John Dawes of the Ehrenberg-Bass Institute for the LinkedIn B2B Institute (2021) — holds that roughly 95% of business buyers are out-of-market at any one time. A film built only to convert the 5% in-market today wastes most of its reach; a film wired to build mental availability and capture the moment a viewer enters the market is an asset. The difference is whether the view can be traced to a contact.

That ratio is also what makes video's dark funnel structurally larger than any other channel's. Paid search meets a buyer at the moment of intent, so the click and the purchase sit weeks apart at most. Video meets a buyer who has no intent yet, and the gap between the view and the enquiry is measured in months or quarters. By the time the buyer acts, the session that carried the film is long expired, the cookie is gone, and the person who watched may not even be the person who fills in the form. The influence did not fail. The instrument did. The 95-5 rule and your video budget works through what that ratio implies for how the money should be split.

Where a corporate video actually gets watched

The abstract version of this problem is easy to nod along to and hard to act on. The concrete version is a list of the six places your last film was watched, and what your stack saw in each. For most B2B teams the honest answer is that one of six is instrumented and the other five are dark.

Where corporate video is watched and what the analytics stack records
Where it is watched	What your stack records
Native social feed	A view count and a watch-time curve, held by the platform. No identity you can act on.
A video platform	Views, retention, sometimes a domain. Rarely a person, and almost never a person your CRM knows.
A page you control	Everything — session, source, behaviour, and an identity if the page asks for one. The only instrumented destination on the list.
A sales deck or proposal	Nothing, unless the file is served from a link you own. Usually the file was emailed.
A forwarded link inside the buying committee	Nothing attributable. The referrer is stripped and the traffic lands as direct.
A screen at an event or in a boardroom	Nothing. This is frequently where the highest-value viewing happens.

Read down the right-hand column and the shape of the problem is obvious. It is not that video cannot be measured. It is that video is consumed in five places designed for reach and one place designed for capture, and the commissioning decision almost never specifies which of the six the asset is actually for.

The useful response is not to stop distributing to the five. Reach is the point of most of them. It is to make sure that at least one canonical, instrumented instance of every asset exists on a page you own, so that the film has somewhere to send the fraction of viewers who want to go further — and so that you have a floor number to report instead of nothing at all.

They are living inside what researchers call the dark funnel — reading, watching, listening, forming opinions, building shortlists in their heads — and leaving almost no traceable signal while they do it.

Paul Joseph Episode 01

The three places attribution breaks

The chain that should carry a viewer to pipeline has three links, and the dark funnel opens wherever one snaps.

01 · Pipeline — the goal

The funnel stage the video serves is never defined, so there is no destination to measure the view against.
02 · Stack — the plumbing

The martech stack cannot attribute a view to a known contact, or the viewer is sent to a generic site with no tracked next step, so intent is captured nowhere.
03 · Craft — the fuel

The brief never tied the asset to a pipeline outcome, so a well-made film is asked to produce a result it was never designed to reach.

Those three links are the scoring dimensions of the Dark-Funnel Video Audit, the diagnostic framework in the Frame to Funnel Method. The order matters: a stack fix applied to an asset with no defined funnel stage produces precise measurement of something nobody agreed the film was for.

A worked example: what the gap costs

Illustrative arithmetic with round numbers, not a benchmark. The point is the shape of the gap and the size of the decision it distorts.

A B2B software company spends $120,000 a year on video: a $40,000 brand film, four customer stories at $12,000 each, and $32,000 of cut-downs, event content and product explainers. Average deal size is $60,000. They close 100 deals a year.

At the annual review, marketing pulls the report. The CRM shows a video touch on 9 of the 100 closed deals — the nine where somebody happened to watch on a page the company controlled and then filled in a form in the same session. Nine deals at $60,000 is $540,000 of influenced revenue against $120,000 of spend. Defensible, but thin enough that finance asks whether the brand film needed to cost $40,000.

Then they add one free-text “how did you hear about us?” field to the demo form, and ask the same question in the first sales call. Two quarters later, buyers on 34 of 100 deals mention video unprompted — the brand film in an investor deck, a customer story a champion forwarded, a conference talk someone watched on a laptop. The stack had captured 9 of those 34.

Evidenced versus buyer-reported video influence across 100 closed deals
Annual video spend	$120,000
Deals closed	100
Video touch recorded by the stack	9 deals · $540,000
Video influence reported by the buyer	34 deals · $2,040,000
Sitting in the dark funnel	25 deals · $1,500,000

Nothing about the video programme changed between those two reports. One instrument was added. The first number was not wrong — it was the floor, presented as though it were the ceiling, which is how a functioning channel gets defunded.

The $1.5m figure is not a claim that video caused those deals; a mention is influence, not causation, and it should be labelled as such in any report. What it is, unambiguously, is 25 deals' worth of evidence that the annual review never saw. And it reframes the question in front of finance. At 9 deals the argument is about whether to keep funding video. At 34 the argument is about which of the six destinations above deserves an instrumented landing page first — a far more productive conversation, and one that only becomes available once the gap has a number.

You do not need to close the whole gap. Halving it roughly doubles the evidenced contribution, which is usually enough to settle the funding question for a year and buy the time to do the rest properly. The full arithmetic — what to divide by what, and how to hold a film to a cost per acquisition — is set out in how to measure video marketing ROI, and the per-asset version runs in the CAC-to-Frame calculator.

Four symptoms of a dark-funnel problem

Before the audit, there are four patterns that reliably indicate the gap is large. Any one of them is enough to start.

Direct traffic is your best-converting channel. It usually is not a channel at all. It is the bucket where every untracked influence lands once the referrer is stripped — including the film someone watched last quarter and the podcast episode a colleague sent.
Sales says video helps and the dashboard says it does not. When the people in the calls and the people in the reports disagree this consistently, the reports are usually measuring a narrower thing than they appear to. Sellers hear the influence; the stack was never wired to record it.
Branded search rises after a campaign, with no attributed conversions. A lift in people searching your name is the clearest available proxy for video doing its job. It is also, by construction, untraceable back to the asset that caused it.
Nobody can name which film influenced a deal. If the CRM records “web” or “video” rather than the specific asset, the reporting cannot inform the next commission — which means every commission is made on taste rather than evidence.

How to trace and fix your own leak

You do not need a new attribution platform to start. You need to test the three links on the video you already have, find the weakest one, and rewire that link before you commission the next asset. The Dark-Funnel Video Audit scores those three links in two minutes and names the single highest-impact fix.

The sequencing rule is worth stating plainly, because getting it wrong is the commonest way a year of attribution work produces nothing. Fix the link the audit names, on one asset, end to end. Not all three links at once, and not the same link across the whole library. An asset that is instrumented from view to contact record teaches you more about where your particular stack breaks than a portfolio-wide policy written before anyone has watched a single view travel the whole chain.

If the weakest link is Craft, the fix happens before production rather than after it: the 60-Second Brief forces the five decisions — funnel stage, viewer action, martech trigger, contact record, pipeline outcome — that a film cannot be held to if the brief never made them. If it is Stack, the instrumentation sequence is in video attribution for B2B.

What you can recover, and what you cannot

Being honest about the ceiling matters, because a programme that promises complete attribution gets abandoned the first time it fails to deliver it. Complete attribution of B2B video is not available to anyone, at any budget. Useful attribution is, and the difference between the two is a reporting decision rather than a tooling one.

Recoverable. Any view that reaches a page you control, any next step the viewer chooses to take, any identity they volunteer, and any answer they give when asked directly. That last one is the most underrated instrument in B2B measurement: a free-text question on the form and the same question in the first call reaches influence no tracking pixel can see, and costs one field.

Not recoverable. The boardroom screening, the clip forwarded inside a committee, the procurement lead who watched a customer story on someone else's laptop and never touched your website until the shortlist was already drawn. These are not edge cases. In enterprise deals they are frequently where the decisive viewing happens, and no amount of instrumentation will reach them.

The practical implication is that the dark funnel is a quantity to be managed and reported, not a bug to be closed. Name it in the report, put a number on it, and show the number falling. A video report that states its own blind spot is far harder to argue with than one that claims a precision it cannot hold — and it is the version that survives a second quarter of scrutiny.

Defending the budget while the gap is still open

Instrumentation takes two quarters. The budget conversation happens next month. Three things hold the line in the meantime, and none of them requires the plumbing to be finished first.

Report two numbers, both labelled. Evidenced influence and reported influence, side by side, with the gap named as the gap. This is more credible than either number alone, because it demonstrates that you know the difference between what you can prove and what you believe — which is precisely the distinction a sceptical CFO is testing for.

Make the gap a target. Commit to a specific reduction — untraced share down from three-quarters to half by the end of next quarter — and report against it. This converts an unfalsifiable claim about brand influence into an operational metric with a due date, which is the form budget conversations accept.

Bring the verbatims. Buyer language, quoted, is evidence of a kind a dashboard cannot produce. Fifteen customers saying in their own words that a specific film mattered will move a budget meeting further than a percentage nobody trusts the derivation of. Keep the answers as raw text rather than mapping them to a picklist at capture — a dropdown returns only the options you already guessed.

The dark funnel, answered

What is the dark funnel?: The dark funnel is the untracked buyer activity between a viewed asset and a known contact. A prospect watches a brand film, reads a post, or hears a podcast, forms an impression, and later arrives through a channel that click-tracking never sees. The influence was real; the attribution was invisible.
Why is corporate video especially prone to the dark funnel?: Corporate video is especially prone to the dark funnel because it is usually commissioned for reach and impression, not capture. A brand film is watched, admired, and shared, but rarely wired to a tracked next step or a martech trigger. Without that plumbing, the view never resolves to a contact, so the spend cannot be traced to pipeline. Video also travels better than any other asset — into a feed, a deck, a forwarded file, a meeting-room screen — and every one of those destinations is somewhere your analytics does not reach.
How do you measure the dark funnel in your own video spend?: You measure it by testing three links: whether each video is commissioned against a defined funnel stage, whether your martech stack can attribute a view to a known contact and a tracked next step, and whether the brief tied the asset to a pipeline outcome. The Dark-Funnel Video Audit scores exactly these three links in two minutes. The rougher version costs nothing: count the deals your CRM shows a video touch on, then count the deals where a buyer mentioned video unprompted. The distance between those two numbers is your dark funnel.
How do you fix a dark-funnel leak?: You fix it by wiring one video end to end before scaling: define the funnel stage first, give the viewer a tracked next step, match the view to a known contact, and fire one automation on view. One instrumented asset teaches the stack more than ten untracked ones.
Can you eliminate the dark funnel completely?: No, and a plan that assumes you can will fail. Some of the most valuable video influence in B2B happens in places no instrumentation reaches — a film played in a boardroom, a clip forwarded inside a buying committee, a conference talk watched on someone else’s laptop. The realistic goal is to shrink the untraced share quarter over quarter and to report the remainder as a named quantity rather than pretending it is zero.
Is the dark funnel the same thing as dark social?: They overlap but are not the same. Dark social describes a distribution problem: sharing that happens over private channels where the referrer is stripped, so the traffic lands as direct. The dark funnel is broader — it covers every untracked influence between exposure and a known contact, including views that generate no click at all. For corporate video the distinction matters, because most video influence produces no session to misattribute in the first place.

Related: defending a video marketing budget covers the rest of the review these three moves are for, video attribution is the operational layer for closing this gap, measuring video marketing ROI covers the arithmetic once it is closed, when a B2B video strategy is not working diagnoses the four failures the dark funnel is only one of, video for demand generation is where the unrecorded views are being generated, the Frame to Funnel glossary defines every term the show uses, and the Corporate Video Buyer’s Guide shows how to commission your next asset on pipeline, not on a showreel. Why brand videos don't generate leads is the same gap read from the film's side, and B2B video marketing is the system that closes it, and does video lower customer acquisition cost? is the question the closed version finally lets you answer.