Field guide

Does video lower customer acquisition cost?

Only when it is attached to something the funnel can see.

By Paul Joseph · Updated 23 Aug 2026

Yes — corporate video can lower customer acquisition cost, and in most B2B companies it currently does not. The mechanism is not mysterious and it is not about production values. Video moves CAC through four specific routes: it raises conversion rate at a step you are already paying traffic for, it shortens the sales cycle, it is a fixed cost amortised over every deal it touches, and it builds the mental availability that makes future demand cheaper to capture. A film that is not attached to a step, a stage or a next action triggers none of those routes. It just adds its cost to the numerator.

The fastest way to find out which describes your own library is the Dark-Funnel Video Audit — two minutes, free, and it names the link that is stopping your video from reaching the denominator.

What CAC is, and where video can touch it

Customer acquisition cost is total sales and marketing cost over a period divided by the number of new customers acquired in it. That is the whole formula, and its simplicity is what makes the video question tractable: there are only two ways to move the number. Reduce the cost, or increase the customers. Video is capable of both, and the confusion in most video business cases comes from arguing for one while measuring the other.

Production cost lands in the numerator immediately and unambiguously. Everything a film might do for the denominator arrives later, partially, and through instruments that were usually never installed. That asymmetry — certain cost now, uncertain benefit later — is the entire reason video budgets are the first to be questioned, and it is a measurement problem long before it is a value problem.

It is also why the honest answer to “does video lower CAC?” is conditional rather than yes or no. The condition is whether anyone built the path from view to contact record. Where that path exists, the four mechanisms below are observable. Where it does not, the same film produces the same influence and none of it reaches the report — which is the dark funnel, and the reason this question is contested at all.

The four mechanisms, in order of how fast they show up

These are not four ways of saying the same thing. They act on different parts of the formula, resolve on different timescales, and require different evidence. Arguing for the slowest one when the fastest is available is the commonest own goal in a video business case.

The practical instruction is to build the case on mechanisms one to three, which you can evidence within two quarters, and to treat mechanism four as the reason not to cut brand-level work rather than as the justification for commissioning it. The 95-5 arithmetic is worked through here, including what it implies for the split between the two kinds of asset.

A worked example: one placement, one quarter

Illustrative arithmetic, not a benchmark. The numbers are round so the mechanism stays visible.

A B2B software company runs paid search to a demo request page. In the baseline quarter they spend $60,000 on paid traffic, which produces 6,000 visits to that page. The page converts at 4% — 240 demo requests. Sales closes 15% of demos, so 36 new customers. Add $30,000 of sales cost attributed to those deals and the acquisition cost is $90,000 ÷ 36 = $2,500 per customer.

They then commission one 90-second explainer for $18,000 and place it above the form on that page. Nothing else changes: same campaigns, same budget, same sales team. The page converts at 5.2% instead of 4% — a 30% lift, which is unremarkable for a well-placed explainer on a considered purchase and well within the range you should expect to argue for.

In the first quarter, with the entire $18,000 production cost charged against it, CAC falls by roughly 8%. That is the pessimistic reading, and it is the one to lead with in front of finance, because it concedes the point they were going to make.

The honest reading is better. The $18,000 is spent once; the 1.2-point conversion lift recurs every quarter the asset stays live and accurate. In quarter two the production cost is gone from the numerator and CAC on the same traffic is $90,000 ÷ 47 = $1,915 — a 23% reduction against baseline, held for as long as the film remains in service. Over eight quarters the explainer's share of acquisition cost is $18,000 ÷ 376 additional customers, or about $48 a customer.

Three cautions, because a business case that ignores them will be dismantled in the meeting. The 15% close rate is assumed constant, and incremental demos from a wider top of funnel often close slightly worse. A conversion lift measured without a holdout is a correlation — run the placement as a split test if your traffic supports it. And the asset decays: a film naming a product tier that no longer exists stops working long before anyone notices it stopped.

Sizing a whole programme rather than one asset is the same calculation one level up — how to set a B2B video marketing budget works it through. Run the per-asset version on your own numbers in the CAC-to-Frame calculator, which does the per-asset version — acquisition cost offset over the asset's life, divided by what it cost to make.

Video against paid media, as cost structures

The question is usually posed as a choice — should this quarter's money go to video or to ads? Posed that way it has no good answer, because the two are not the same kind of cost and do not compete for the same job.

Paid media is variable and instantly responsive. Spend more, get more, stop spending and it stops within the day. Its cost per acquisition is roughly flat and tends to rise as you scale, because you exhaust the cheapest inventory first. Video is fixed and slow. It costs the same whether ten people or ten thousand watch it, which means its cost per acquisition falls monotonically as long as the asset is in service.

Those two shapes are complements, not substitutes. The most dependable CAC effect available to a B2B marketing team is not replacing paid media with video — it is placing video inside the paid funnel so the traffic already being bought converts better, which is exactly the worked example above. The film does not need to generate its own audience to pay for itself; it needs to improve the economics of the audience you are already buying.

This also disposes of the framing that video is expensive. A $18,000 explainer is about a fortnight of that company's paid search spend. The difference is that at the end of the fortnight the ads have stopped and the explainer has not, and nobody describes the recurring cost as the expensive one because it never arrives as a single invoice.

Pricing the cycle-time effect

Mechanism two is the one most often asserted and least often priced, which is a shame, because selling time is usually the largest variable cost in B2B acquisition and it is entirely visible in a CRM. The arithmetic is straightforward once you decide to do it.

Take a seller on a fully loaded cost of $150,000 a year. Assume 1,600 productive hours, and the cost of an hour of selling time is roughly $94. In the company from the worked example above, a deal takes an average of eleven seller-hours across discovery, two technical calls, a security review and a negotiation. That is about $1,030 of selling cost per closed deal before any marketing spend is counted, and it recurs on every deal, forever.

Now suppose a $12,000 customer story removes one call from the middle of that sequence — the one where a seller talks a sceptical technical buyer through how a comparable business implemented the product. Sales estimate that call at ninety minutes including preparation and follow-up, and report that prospects who watched the film arrive at the next call having already asked the questions it answered.

The asset pays for itself inside a year on cycle time alone, before counting a single incremental deal. That is the version of the business case worth carrying into a planning meeting, because it does not depend on a conversion lift anybody can dispute — it depends on a call that either happens or does not, which the sales team can confirm.

There is a second-order effect that is real and harder to defend, so mention it and do not lean on it: the hours released are hours a seller spends on another deal. Whether that converts into more revenue depends on whether the constraint on your growth is seller capacity or demand. If it is capacity, the cycle-time mechanism is worth considerably more than the table shows. If it is demand, it is worth exactly what the table shows, and the conversion mechanism is where the leverage sits instead.

The instrumentation this needs is lighter than it sounds. You do not have to time every call. You need the asset recorded against the deals it touched — decision four of the 60-Second Brief — and a comparison of average cycle length between deals that touched it and deals that did not. Two CRM reports, run quarterly.

When video raises CAC

A page arguing the affirmative case is worth nothing unless it also states the conditions under which the answer is no. There are four, and they account for most of the video spend that ends up defended by adjectives.

• No funnel stage was specified. An asset commissioned without a stage cannot be placed at a step, so mechanism one never fires. This is the most common failure and the cheapest to prevent: it is decision one of the 60-Second Brief.
• The view cannot reach a contact record. If the only place the film lives is a channel where a view resolves to nothing, the influence is real and permanently unevidenced. Fixing that is instrumentation, not production — see video attribution for B2B.
• The asset is replaced on a calendar. Remaking a film annually because it feels stale resets amortisation to zero every year. Mechanism three is the one that does the heaviest lifting in the arithmetic, and it is the one most often destroyed by a rebrand nobody costed against acquisition.
• The budget was set by ambition, not by the offset. If the production cost was never checked against the acquisition cost the asset could plausibly offset, the film can be excellent and still be the wrong size. What corporate video costs sets out what actually drives the number.

A fifth condition is worth naming because it is the only one that is not a mistake. Some assets are commissioned for a job that has nothing to do with acquisition — a recruitment film, an investor piece, a safety induction. Holding those to a CAC contribution is a category error, and the right response is to say so in the brief rather than to construct an acquisition argument nobody believes. A budget that distinguishes the two is far easier to defend than one that pretends every asset is a growth investment.

What to do with this in the next planning cycle

The instruction that follows from all of the above is narrower than it looks, and it is deliberately unambitious: stop asking whether video lowers CAC in general, and instrument one asset well enough to answer it for your own business. A single instrumented film produces a defensible number. A library of uninstrumented ones produces an argument, and arguments lose to spreadsheets.

Concretely: pick the step in your funnel with the most traffic and the worst conversion rate, because that is where mechanism one has the most room to work. Commission one asset against that step, using the Frame to Funnel Method to name the stage, the viewer action and the trigger before anything is shot. Wire the placement so a view can reach a contact record. Then hold the baseline for a quarter and measure.

That gives you one number, from your own funnel, with your own close rate, that nobody in the room can wave away. It is worth more than any benchmark, including the ones on this page — and once you have it, the next commission is a sizing decision rather than a fight.

Related: the CAC-to-Frame Ratio runs this arithmetic per asset, how to measure video marketing ROI covers the wider measurement method, and the glossary defines every term used here.