How to Measure Video ROI: The Business of Creative Assets

By Chris Fischer, Director of Business Development, Rushlight Agency

Ask most marketing directors what a video costs, and they have an instant answer. Ask if the video worked, and their response will usually get vague. This is because cost and return are two different questions, and most marketing conversations only answer the first. Of course, reach matters, but it should not be considered the final goal. We like to say that every creative asset is a math problem hidden under a creative one, and measuring video ROI is simple once you know how to track the right numbers.

Metrics That Feel Like Answers But Aren't

Views, followers, and engagement rate all describe attention. None of them describe money. A video with 50,000 views and a video with 500 views can produce the exact same number of sales, or the smaller one can produce more, because view count was never actually measuring the thing that matters.  This blog series (HERE) has already made that case from the audience's side: that basic following or views proves attention, not revenue. This blog lays out the other half of that argument: how to measure ROI.

The Three Calculations That Do the Real Work

Cost per view (CPV): total cost divided by total views. A video that cost $8,000 to produce and distribute, generating 40,000 views, comes out to $0.20 per view. That number alone means nothing. There's no universal "good" CPV number, only one that justifies the cost of the view itself.

Cost per acquisition (CPA): total cost divided by the number of customers the asset actually generated, tracked through a real, specific link, promo code, or attribution point, not estimated after the fact. If that same $8,000 video generated 12 new customers, CPA lands around $667. Whether that's a win depends entirely on the next number.

LTV:CAC ratio: what a customer is actually worth over time, compared to what it cost to acquire them. David Skok, a venture capital investor at Matrix Partners, laid out the standard version of this guideline in 2013, in an essay called "SaaS Metrics 2.0," based on years of working with software companies. he best businesses keep a ratio higher than 3, meaning three dollars of customer value for every dollar spent acquiring that customer, with the strongest performers reaching 7 or 8. Skok is careful to call it a guideline, not a rule, and he notes plenty of healthy young businesses haven't hit it yet, but can see how to get there over time. The logic still transfers outside software: if a customer is worth $150 over their relationship with a business and it cost $667 to acquire them through that video, the video lost money, regardless of the view count. If that same customer is worth $3,000, the video was a genuinely excellent investment, and the view count wouldn’t matter.

Why a Video Isn't a One-Time Expense

Treating an $8,000 video as a cost that either paid off or didn't in its first week is where a lot of this math breaks before it even starts. Wistia's 2026 research on video content found that almost 90 percent of companies now take one longer video and cut it into months of shorter clips for social, email, and ads. This means the true cost of that asset is spread across everything it becomes, not just its first iteration and showing. Accountants have a name for this logic: amortizing an expense, recognizing its value over the period it actually produces value, instead of expensing the whole cost the day it was made. A video used once is an $8,000 expense against whatever it produced that week. The same video, repurposed into a dozen pieces of content over six months, is a very different number once the math is actually run. The CPA and LTV:CAC figures change completely depending on which version of that math is being done.

The Time Variable: How Long It Works Matters as Much as What It Cost

None of the three calculations above ask the question that actually separates a documentary-style brand film from basic social content: how long does the asset keep working after it's made? A single, higher-priced brand film can sit on a website, in a sales deck, and in a pitch for years, still producing leads with zero additional spend. A basic piece of social content is often functionally gone within days, buried by a feed that was never built to resurface anything old, which means producing more of it isn't optional, it's the price of staying visible.

Scott Graffius's annually updated study, drawing on more than 5.6 million posts analyzed for its 2026 edition, found the median time for a post to reach half its total lifetime engagement runs from close to zero minutes on TikTok and Snapchat, to 52 minutes on X, to about 18 hours on Instagram, out to roughly 10.6 days on YouTube and around 2 years on a blog post. Engagement half-life isn't the same thing as sales half-life, but it's the closest verifiable proxy available for how long a piece of content actually stays in front of anyone.  The pattern matches what anyone running a content calendar already knows by feel: short-form social decays in hours, long-form video and written content decays in months or years.

That decay rate changes the math. A cost-per-acquisition calculated only in the first week after launch is a snapshot, not the full picture, and the snapshot and the full picture can tell very different stories about the same two projects. Call the fuller version effective CPA: total cost over an asset's full functional life, including any recurring spend required to keep producing it, divided by total acquisitions over that same life.

Here's why that split matters. A $25,000 brand documentary that keeps generating 5 qualified leads a month for 3 years, with no further production spend required, produces 180 leads for a final effective CPA of about $139. A recurring social content program costing $2,200 a month to keep staffed and produced, generating 3 leads a month, while the spend continues, produces 108 leads over that same 3 years for a total spend of $79,200, an effective CPA of about $733. The social program had the lower price tag on day one. It did not have the lower cost per lead over time, because it never stopped needing new spend to keep producing, while the documentary kept paying out long after the expense was realized.

That doesn't make recurring content the wrong choice. A business that needs to stay visible weekly, respond to trends, or support a sales team's daily posting can't get that from one film made three years ago, no matter how well it performs. The point isn't that evergreen always wins. It's that a lower sticker price with a short functional life and a recurring bill attached isn't automatically the cheaper option, and a higher sticker price with a long functional life and no recurring bill isn't automatically the more expensive one. Which one actually costs less per result only shows up once time is put into the calculation, not left out of it.

The Real Question

None of this requires new software or a data analyst. It requires tracking four things a lot of marketing conversations skip entirely: what something actually cost, including the time to make it and the money to distribute it; what it actually produced, tracked through something specific, not estimated after the fact; what that customer is actually worth, which most businesses already know or can find in a few minutes of looking at repeat purchase data; and how long the thing actually keeps producing anything before it needs to be remade or replaced. Run those four numbers and the question stops being whether people watched something and becomes whether it made more than it cost, for as long as it actually ran. For a business trying to make a thoughtful decision about its next round of creative spend, that's the only question in the business of creative assets actually worth asking.

A Practical Worksheet

Running this math for the first time is easier with the numbers laid out in front of you instead of built from scratch. Rushlight put together a one-page worksheet that walks through exactly this calculation: what a piece of content cost, what it produced, and whether the return actually cleared the bar. It's built to be filled out for any single video, campaign, or piece of content, and it's free to use, whether or not that content came from us. Learning how to measure video ROI doesn't require a finance degree. It requires asking the question at all.

Chris Fischer is Director of Business Development at Rushlight Agency in Salem, Oregon. chris@rushlightagency.com