How to measure video marketing ROI

To measure video marketing ROI, subtract the total cost of the video from the value it generated, then divide by that cost. Value means one specific business outcome you chose before shooting: pipeline created, sales assisted, trials started, support tickets avoided, or media spend saved. Cost means everything, including internal hours. The reason so few teams do this is not the maths, it is that the outcome was never named, so at the end there is nothing to measure against except views.

Views are not a result. They are the price of admission, the number you have to clear before anything else can happen, and they tell you almost nothing about whether the video was worth making. This guide covers the actual formula, the three layers of metrics that matter and which to use when, the two mistakes that make most video ROI numbers wrong, how to set the target before you shoot, and a measurement plan you can fill in today.

How do you calculate video marketing ROI?

Use the standard ROI formula and be honest about both halves of it:

Total cost is where most calculations quietly cheat. The invoice from the production company is the visible number, but for in-house teams the larger cost is time: the hours spent briefing, reviewing, re-reviewing, and cutting versions for four platforms. If a marketer, a designer, and a product manager each spend a day on a video, that is a real cost and it belongs in the denominator. Leaving it out does not make your video look better, it just makes your ROI number untrue in a way that will not survive a finance conversation.

Value generated is where you have to make a choice, and it should be made in the brief rather than after the fact. Pick one:

• Pipeline or revenue assisted. Best when the video sits in a considered buying journey and you can see which deals touched it.
• Conversions or trials started. Best for a product video on a page where you can run the page with and without it.
• Cost avoided. A support explainer that deflects tickets, or an internal video that replaces a recurring live session, both have a value you can calculate from time saved.
• Media efficiency. If a new creative drops your cost per acquisition on the same spend, the saving is the return.

Which metrics actually tell you if a video worked?

Three layers, in order, and each one is only useful if the layer above it cleared. Reach tells you the video was distributed, attention tells you the creative held, and outcome tells you the business got something. A video can win the first two and still be worthless, which is exactly why teams that stop at layer two end up unable to defend a budget.

The layer teams reach for is shifting, and not always toward the useful end. Social engagement has become the fastest-rising success metric in video marketing, which is a layer-two measure being asked to do a layer-three job.

Engagement is a perfectly good metric as long as you know which question it answers. The problem starts when it becomes the only number in the report, because then a video that was briefed to generate demos gets judged on comments, and nobody notices that it generated no demos.

Why do most video ROI numbers come out wrong?

Two reasons, and both are structural rather than analytical. The first is that the video data never reaches the systems where outcomes live.

This is a plumbing problem wearing the costume of an analytics problem. If your video tool knows someone watched 90% of a demo and your CRM knows that same person booked a call, but the two never exchange that fact, then you will report the view and lose the outcome. Connecting them is usually an afternoon of setup, and it changes what you are able to claim for every video you make afterwards.

The second reason is timing. Most teams close the measurement window a few weeks after publishing, which captures the immediate response and misses the rest of it.

Read that carefully, because it reframes the whole exercise: measuring for four months and stopping captures roughly half the return. A video judged at week three is being judged on a fraction of what it will eventually do, and the videos that suffer most from an early verdict are the ones doing brand work rather than direct response.

"Previously, we had considered those two campaign types separately. But the new findings were illuminating." — Sarah Barron, CMO U.K. & Ireland, Domino's

The practical version for a small team: keep a longer window than feels natural, and do not retire a video from reporting the moment the campaign ends. Check it again at 90 days and at six months, especially anything evergreen sitting on a product page or a YouTube channel.

How do you set a video ROI target before you shoot?

Put one number in the brief and make it specific enough to fail. "Increase awareness" cannot be measured, so it cannot be defended. "Lift the pricing page conversion rate from 3.1% to 3.6% within 90 days" can be, and it also tells the editor what the video is for, which usually improves the video itself.

Three questions settle it:

• What number already exists? Measure against something the business already tracks. Inventing a new metric for a video is how a result becomes unarguable in the wrong direction.
• What is the baseline? Write down what that number is today, before the video exists. A result without a baseline is an anecdote.
• What is the window? Pick a date to check, and a second, later date to check again. Committing to both up front stops a slow winner being written off at week two.

This belongs in the brief itself, next to the audience and the message. If you are building briefs from scratch, the structure is covered in how to write a video brief, and the measurement line sits alongside the objective.

What does a video measurement plan look like (example)?

Here is a prompt you can paste into PlanThatVideo for a product page explainer, and the measurement plan a team would attach to it before the shoot.

And the measurement plan that turns it from a video into something you can actually judge:

Notice what this makes possible. At day 30 the drop-off curve tells you whether people are leaving at the jargon-heavy second beat, which is a fixable creative problem you can re-cut in an afternoon. At day 90 you have a number to compare against a baseline you wrote down before you started. And because break-even is 11 trials, the conversation with finance is about whether the video cleared a low, specific bar rather than about whether video is worth doing in general.

A reusable video ROI measurement template

Copy this into the bottom of every video brief. If you cannot fill a line, that is the thing to resolve before booking a shoot day.

How does repurposing change the ROI calculation?

It changes the denominator, and it is usually the fastest way to improve a video's return. The cost of a shoot is largely fixed: the same setup, crew, and travel produce one asset or ten depending only on what you planned to capture. If a single shoot day yields a hero video, four social cuts, and a set of stills, the cost per asset falls by an order of magnitude while the value side keeps adding up.

That is why the planning stage, not the edit, is where video ROI is mostly decided. Capturing the extra angles costs minutes on the day and is impossible to add afterwards. The mechanics of getting ten assets out of one shoot are covered in how to repurpose video content, and scheduling them so the shoots stack is covered in the video content calendar.

FAQ