A company selling field service management software to industrial equipment makers spent $110,000 producing a two-minute brand film. The film was built like a television commercial, with aerial photography and an original score. The media team ran the film as a skippable in-stream ad on the main video platforms, and the dashboard reported 1.6 million views for the quarter. A deeper report showed where those views went. Seventy-eight per cent of viewers skipped within the first six seconds, and the company name did not appear until the final fifteen seconds. Most of the budget had bought a few seconds of high-quality footage that never showed the company's name. The film was excellent. The buy built almost no memory of the company.

Video is the strongest demand creation format in B2B, and video is also the easiest format to waste. This post explains what video advertising is for, how video is bought, where video fits beside the other channel families, how to measure video, and the four ways video reporting overstates the channel. The series hub, How B2B Media Actually Works (and What It Cannot Do), covers the research behind the plan as a whole: the 95:5 rule, the buying committee, and demand creation versus demand capture.

What Video Advertising Is For

Video is a demand creation channel. Among the demand creation formats, video builds the most memory per exposure. The Ehrenberg-Bass Institute's research on B2B buying holds that around 95 per cent of category buyers are not in market in any given period. Those out-of-market buyers cannot be converted this quarter. Those out-of-market buyers can be made to remember the company, so that the company comes to mind when a buying trigger fires months from now. Memory is built through exposure, and formats differ in how much memory each exposure builds. A banner is glanced at. A static feed post is read for a second or two. Video carries sight, sound and motion at once, so a watched video builds more memory per exposure than any static format can. Our post on mental availability in marketing sets out the memory mechanism in full.

Video also does work no static format can do: demonstrating the product, putting a human voice on the company, and explaining a point of view at depth. Gartner's research on the B2B buying journey describes a months-long looping process, run by a buying committee of six to ten people who do most of their research before any supplier hears from them. A video watched by three members of that buying committee this quarter is doing work the plan will only see quarters later, and mostly will never see credited to video at all.

How Video Advertising Is Bought

B2B video runs in three kinds of place. In-stream ads play before or during video content the viewer chose to watch, on the main video platforms. Feed video runs inside the scrolling feeds of the professional social networks, where the video usually auto-plays: playback starts on its own as the post scrolls into view, muted, without the viewer pressing play. The third place is connected TV, which is streaming television delivered over the internet to the living-room screen, with ad slots sold by audience rather than by programme schedule. Pricing takes two main forms. Buying per thousand impressions means paying a fixed rate for each thousand times the ad is served, whether or not anyone watches. Buying per completed view means paying only when a viewer watches the ad to the end or past a set number of seconds.

In-stream formats divide into skippable and non-skippable. A skippable ad shows a skip button after the first few seconds, and the viewer can dismiss the ad and move on to the chosen video. A non-skippable ad plays in full before the chosen video starts, and non-skippable slots are usually capped at a short length. Skippable and auto-play formats together mean that a large share of viewers see only the opening seconds. The opening therefore carries the message. Show the company name and the point of the ad before the skip button appears, and write the opening so the message survives with the sound off, because captions are all a muted feed viewer receives. A video that saves the company name for the closing seconds is a video most viewers never connect to the company.

The barrier that keeps video out of most B2B plans is production cost rather than media cost. A television-grade shoot runs to six figures, so many teams either skip video entirely or produce one expensive film and expect the film to serve every purpose for years. Neither choice is necessary. Simple, message-led video shot for the format beats expensive television-style production in B2B, because the formats reward clarity in the first seconds and punish films that bury the company name until the end. A subject-matter expert on camera, a sharp claim in the first five seconds, a caption track and the logo on screen from the start can be produced in days. The course Media and Measurement carries the method for briefing and structuring video creative by format.

Connected TV deserves a separate note, because connected TV behaves unlike the rest of the channel. The living-room screen reaches decision makers as people, relaxed in the evening, rather than as job titles at a desk. Targeting is by household and by broad audience segment, so firmographic accuracy is weak: the buy will reach some of the right people and pay for many of the wrong ones. There is also no click at all. Nobody taps a television ad, so every effect a connected TV buy produces arrives indirectly, through branded search, direct traffic and recall. Treat connected TV as the purest brand format in the plan, and judge connected TV only on the measures a brand format can move.

Where Video Advertising Fits in the Plan

Video sits inside the demand creation budget beside paid social and programmatic display, and the three formats trade memory per exposure against cost per exposure. Display, covered in Programmatic Display in B2B: Cheap Reach, Rarely Clicked, Often Misjudged, delivers a glance at the lowest price. Paid social, covered in Paid Social in B2B: Reaching Buyers Who Are Not Looking, delivers a second or two of feed attention at a mid price. Video delivers the most memory per exposure at the highest price. Run the formats with distinct jobs: video plants the message with the richest creative, paid social repeats the message in the feed, and display refreshes the message cheaply between exposures.

Video also feeds retargeting. The platforms build audiences of people who watched a chosen share of a video, and those video-engaged audiences can then receive follow-up messages across social, display and video inventory. In my experience, video earns a growing share of the demand creation budget once two conditions hold: the company can produce format-fit creative without a six-figure shoot, and the company can commit to running video for several quarters rather than testing video for six weeks. A single burst of video builds memory that decays before most buying triggers fire.

How to Measure Video Advertising

Measure video the way demand creation is measured. In quarter: reach and frequency within the ideal customer profile (the defined set of companies and roles the company sells to), and completed views within that profile. Those delivery readings answer whether the right people watched. Over time: aided awareness and recall, measured by survey. Aided awareness is the share of surveyed buyers who recognise the company when shown a list of suppliers, and recall is the share who can bring the company or the message to mind unprompted. As leading indicators: lift in branded search volume and in direct traffic in the markets where the video ran. On a lag: pipeline among reached accounts, compared with matched accounts the video did not reach, read two to four quarters after the spend. Binet and Field's analysis of the IPA databank explains the lag: brand-building effects compound slowly and persist, while activation effects spike and fade, so a video programme judged inside one quarter will always look like waste.

A completed view is the strongest delivery signal video offers, and a completed view can still be worthless. Completion only means the ad reached the end while counted as playing. On cheap auto-play placements, a muted video can play to completion in the corner of a screen nobody is watching. A completion earned on a skippable placement carries real information, because the viewer held the power to skip and chose to stay. A completion manufactured on inventory the viewer could not skip or never noticed carries none. What platform reporting can never provide is proof of cause, and why marketing attribution misleads covers the reasons.

The table below is a worked example for a company selling workforce management software to large retailers. The company runs two video buys at the same quarterly spend. The first buy is optimised to cost per view on the cheapest in-stream and auto-play inventory available. The second buy is restricted to skippable in-stream placements and professional network feed placements, served only within the ideal customer profile.

Example figures for illustration. Engaged target accounts: accounts on the target list showing new website visits or sales conversations within two quarters of the campaign start. Branded search lift is directional: compared against the quarter before the campaign, with no control group.

The cheap buy wins the view count by a factor of six, and the view count is the only reading the cheap buy wins. Restricting the second buy to chosen formats within the ideal customer profile cut the raw views and multiplied the completed views among people the company sells to. Cost per completed view within the profile, branded search lift and engaged target accounts all favour the second buy. Any report ranking the two buys on views counted would move the budget onto the inventory nobody watched.

Where the Numbers Lie

Video reporting overstates the channel in four recurring ways.

1. View counts are reported without saying what counts as a view

A view sounds like a person watching. In reporting, a view is whatever playback threshold the platform serving the ad chose to apply. Some platforms count a view after two or three seconds of playback, muted and auto-playing, while others count at longer thresholds or at completion. Two consequences follow. View totals from different platforms cannot be added together or compared, because the totals count different events. And a large view total mostly does not represent watched video, because the lowest thresholds are met by ads that scrolled past in a feed. What to do: attach the counting definition to every view figure in every report, and evaluate the buy on completed views within the ideal customer profile rather than on views.

2. Completion rate is treated as a measure of attention

Completion rate looks like proof that viewers chose to keep watching. On skippable inventory, a completion does record a viewer who chose to keep watching. On non-skippable and auto-play inventory, completion is manufactured: the viewer could not skip, or the video played muted in a feed while the viewer read something else, so the completion arrived without a decision by anyone. Manufactured completions concentrate on cheap inventory, because the placements nobody pays attention to are also the placements other advertisers avoid, which keeps the price of those placements low. A buy can therefore raise its completion rate by moving spend onto worse placements. What to do: read completion rate by placement type, and only credit completion on formats the viewer could have skipped.

3. View-through conversions claim pipeline that other channels closed

A view-through conversion is recorded when someone converts after being served a video impression, with no click involved. Video buys serve impressions at large volume, so almost every eventual buyer has a video impression somewhere in the trail. The video report then claims conversions that paid search intercepted, the website persuaded and the sales team closed. The claim scales with impression volume rather than with influence, so the cheapest and least attentive buys claim the most credit. Treat a view-through conversion as a record of exposure, never as a record of cause. Where the claimed credit would move budget, test the claim by comparing reached accounts with comparable accounts held out of the campaign.

4. Cost per view optimisation steers budget to the least attentive placements

Optimising a video buy to cost per view instructs the auction to find the cheapest countable views available. The cheapest countable views are the least attended: auto-play slots, low-quality app inventory, and placements where a muted video plays to nobody. Reported efficiency improves every week while the buy's real output, which is memory among buyers, falls. The mechanism is the same mechanism that ruins display buys optimised to clicks, covered in Programmatic Display in B2B: Cheap Reach, Rarely Clicked, Often Misjudged. What to do: optimise to completed views or reach within the ideal customer profile, cap the share of auto-play inventory in the buy, and read the placement report every month.

What Video Advertising Cannot Do

Video cannot capture demand, because nobody summons a video ad the way a search query summons a search ad. Video cannot produce leads this quarter at an acceptable cost, and a video campaign judged on demo requests will always look like a failure, because demo requests are demand capture and video does demand creation. Video cannot rescue an unclear message: production polish adds nothing when the opening seconds fail to say who the company is and what the company claims. Video cannot work as a single burst, because the memory one burst builds decays before most buying triggers fire. Connected TV cannot target firmographics with precision and cannot be clicked, so connected TV cannot feed retargeting audiences or platform conversion reports. Video cannot prove its own contribution from platform reporting, because views are defined by the platforms and view-through conversions are claims. What video can do is build more memory per exposure than any other format in the plan, among the buyers who will not enter the market until next year.

Video Advertising FAQs

What counts as a view in video advertising?

Whatever the platform serving the ad decides. Some platforms count a view after two or three seconds of muted auto-play, while others count at longer thresholds or only at completion. A view total is therefore meaningless without the definition attached, and totals from different platforms cannot be compared. Ask for the definition behind every view figure, and evaluate on completed views within the ideal customer profile instead.

Should B2B video ads be skippable or non-skippable?

Skippable placements produce the more trustworthy signal, because a completion on a skippable placement records a viewer who chose to stay. Non-skippable placements guarantee that a short message is delivered in full, and non-skippable placements also manufacture completions that prove nothing about attention. Both formats can earn a place in a buy. Judge the creative on skippable completion rates, and never let non-skippable completions inflate the report.

How much does B2B video production need to cost?

Far less than most teams assume. The formats reward a clear claim delivered in the first seconds, a caption track, and the company name on screen from the start. A subject-matter expert on camera with a sharp message can be filmed and edited in days. Spend on message clarity and on enough variations to avoid wear-out before spending on cinematic polish, because polish without an early, clear message buys admiration for footage nobody connects to the company.

Is connected TV worth testing in B2B?

Worth testing when the category is broad enough that household-level targeting waste is tolerable, and when the company already measures branded search and direct traffic well enough to detect lift. Connected TV reaches decision makers as people on the living-room screen, with weak firmographic accuracy and no click at all, so every effect arrives indirectly. Run connected TV in defined regions, compare branded search and direct traffic against unexposed regions, and expect the reading to take quarters.

How long does video advertising take to show results?

Delivery readings arrive immediately: reach, frequency and completed views within the ideal customer profile confirm the right people are watching. Leading indicators, such as lift in branded search and direct traffic and newly engaged target accounts, move over one to two quarters. Pipeline among reached accounts arrives two to four quarters after the spend, in step with the buying cycle. Binet and Field's IPA databank analysis shows the delay is structural: brand-building effects compound slowly, and short evaluation windows systematically undercount brand-building effects.

Go Deeper

This post carries the judgement: what video builds, why the opening seconds decide the outcome, and why views and completions need careful reading. The method sits in the course Media and Measurement, which covers format choice, creative briefs for the first seconds, view definitions and account-level measurement step by step. The free module in B2B Marketing Fundamentals covers the ideal customer profile that every video audience inherits. The memory mechanism video serves better than any other format is the subject of mental availability in marketing.

Sources

• Ehrenberg-Bass Institute for Marketing Science, 2021, How B2B Brands Grow: the finding that around 95 per cent of category buyers are out of market in any given period, and the role of memory and mental availability in which suppliers are considered when buying begins.
• Les Binet and Peter Field, The Long and the Short of It, IPA, 2013: analysis of the IPA effectiveness databank showing that activation effects spike and decay within weeks while brand-building effects compound over years, and that short evaluation windows shift budgets toward activation.
• Gartner, The B2B Buying Journey: typical buying groups of 6 to 10 decision makers, a long non-linear journey that loops through research and evaluation, and most buying time spent away from suppliers.