Video KPIs That Connect to Business Outcomes
Video KPIs are the metrics you use to measure whether your video content is working: views, watch time, completion rate, click-through rate, and downstream conversions. The problem is that most video reporting stops at the surface, counting impressions and completions while staying carefully silent on whether any of it moved the business forward.
Getting video measurement right means knowing which metrics correspond to which business objectives, and being honest about the gap between what platforms tell you and what your business actually needs to know.
Key Takeaways
- Completion rate and watch time are directionally useful, but neither confirms that your message landed or that a viewer moved closer to buying.
- Platform-reported view counts are not standardised. A “view” on YouTube, Meta, and LinkedIn means three different things. Compare them only within each platform.
- The most valuable video KPIs are the ones connected to a business objective. If you cannot draw a line from the metric to revenue, pipeline, or retention, it is a vanity metric until proven otherwise.
- Measurement infrastructure needs to be built before the campaign launches, not retrofitted after the results disappoint.
- Attribution for video is genuinely hard. The answer is honest approximation, not false precision or giving up entirely.
In This Article
Why Video Measurement Is Still a Mess
I have sat in more video campaign reviews than I care to count where the headline number was total views. The client nods. The agency smiles. Nobody asks the uncomfortable question: views of what, by whom, for how long, and what happened next?
The issue is structural. Video platforms have a commercial incentive to make their numbers look good. Meta counts a view at three seconds. YouTube counts it at thirty. LinkedIn has its own threshold. When you aggregate across channels and report “total video views,” you are combining apples, oranges, and things that are not fruit at all. The number is technically accurate and practically meaningless.
This is not a new problem. When I was running paid media at scale, managing hundreds of millions in ad spend across multiple markets, the tension between platform-reported metrics and business outcomes was constant. Platforms optimise for the metrics they report. If you let them, they will optimise your campaigns toward the numbers that make their dashboards look impressive, not the numbers that make your P&L look impressive. Those two things are not always the same.
Good video measurement starts with a clear-eyed view of what each metric actually tells you, and what it does not.
The Core Video KPIs and What They Actually Mean
There is a standard set of video metrics that every platform reports. Here is an honest assessment of what each one is worth.
View Count
The most reported and least useful metric in isolation. View count tells you how many times your video was triggered to play, subject to whatever threshold the platform uses. It says nothing about whether anyone watched, what they took away, or whether they were even the right audience. Use it as a reach indicator only, and always qualify it by platform.
Watch Time and Average View Duration
More useful than raw views because they indicate engagement depth. If your average view duration is 8 seconds on a 60-second video, something is wrong: the creative, the targeting, or the placement. Watch time is also a factor in how platforms distribute content organically, which gives it a secondary value beyond measurement.
The limitation is that watch time does not tell you whether the right message was communicated. Someone can watch 45 seconds of your video and retain nothing of commercial value. Duration is a proxy for attention, not comprehension or intent.
Video Completion Rate
Completion rate is the percentage of viewers who watched your video to the end, or to a defined threshold such as 75% or 100%. It is one of the more reliable engagement indicators because it requires sustained attention. A completion rate that is significantly below benchmarks for your format and platform is a strong signal that the creative is not holding attention.
Where it gets complicated is with longer-form content. A 60% completion rate on a two-minute video represents a very different level of engagement than 60% on a fifteen-second pre-roll. Always contextualise completion rate against video length and format.
Click-Through Rate
Click-through rate (CTR) measures the percentage of viewers who clicked on a call to action, overlay, or end card. It is the bridge between video engagement and downstream behaviour, which makes it one of the more commercially meaningful video KPIs. Low CTR on a high-completion video suggests the creative is engaging but the call to action is weak, misaligned with audience intent, or simply not compelling enough.
CTR varies enormously by placement and format. In-stream ads with forced views will almost always produce lower CTR than content someone chose to watch. Do not compare CTR across fundamentally different placements without accounting for that context.
View-Through Rate and Assisted Conversions
View-through rate (VTR) measures how many people who saw your video went on to convert without clicking. It is where video measurement gets genuinely contested. Platforms love VTR because it attributes value to impressions that produced no direct click. Sceptics argue it conflates correlation with causation and inflates the apparent contribution of video.
Both positions have merit. Video does influence purchase decisions that get attributed to other channels later in the funnel. But platform-reported VTR windows are often generous enough to capture conversions that would have happened anyway. The pragmatic approach is to use VTR as one signal among several, weight it conservatively, and triangulate it against incremental lift data where possible.
If you want to go deeper on how to structure your analytics setup to capture these signals properly, the Marketing Analytics hub covers the infrastructure side in detail, including GA4 configuration and custom reporting frameworks.
Matching Video KPIs to Campaign Objectives
The single biggest mistake in video measurement is using the same KPIs regardless of what the campaign is trying to achieve. A brand awareness campaign and a direct response campaign have fundamentally different success criteria. Measuring them identically produces misleading results in both directions.
Here is how to align metrics to objectives.
Brand Awareness and Reach
Primary KPIs: unique reach, frequency, brand recall lift, share of voice. Secondary KPIs: completion rate, view-through rate. What to deprioritise: CTR and direct conversions. Brand campaigns are not supposed to drive immediate clicks. Judging them on CTR is like judging a billboard on how many people pulled over to write down the number.
Consideration and Engagement
Primary KPIs: watch time, completion rate, engagement rate (comments, shares, saves), website traffic from video. Secondary KPIs: return visitor rate, time on site post-click. This is the middle of the funnel, where video does some of its most valuable work and gets the least credit because it does not convert immediately.
Direct Response and Conversion
Primary KPIs: CTR, cost per click, cost per acquisition, return on ad spend (ROAS). Secondary KPIs: landing page conversion rate, assisted conversions. This is where video measurement is most straightforward because you have a clear outcome to attribute against. The challenge is ensuring your attribution model is not over-crediting the last click and under-crediting the video that drove the intent.
I saw this play out clearly at lastminute.com, where a relatively simple paid campaign for a music festival drove six figures of revenue in roughly a day. The video and display elements that had been running in the weeks before barely got credit in the last-click model. The search campaign that captured the final intent looked like a genius move. It was a good campaign, but it was standing on the shoulders of awareness work that the reporting largely ignored.
Building a Video Measurement Framework
A measurement framework is not a spreadsheet of metrics. It is a structured approach to deciding what you will measure, how you will measure it, what good looks like, and what you will do when the numbers tell you something is wrong.
Step 1: Define the Business Objective First
Before you open a platform dashboard, write down the business outcome this campaign is supposed to support. Not the marketing objective. The business objective. Revenue, pipeline, customer retention, market share. Everything else flows from that. If you cannot connect your video KPIs to a business outcome, you are measuring activity, not performance.
Step 2: Set Up Tracking Before the Campaign Launches
This sounds obvious. It is routinely ignored. UTM parameters on every video link. GA4 events configured to capture the downstream actions that matter. Conversion tracking verified and tested before spend goes live. GA4 has capabilities that most teams are not using, including event-based tracking that can capture video engagement milestones and connect them to conversion paths.
Retrofitting tracking after a campaign has run is one of the most frustrating exercises in marketing. You end up with platform data on one side, business data on the other, and no reliable way to connect them. I have been in that position. It is avoidable.
Step 3: Establish Benchmarks
A completion rate of 45% means nothing without context. Is that good for your format, your industry, your audience? Build benchmarks from your own historical data first. Supplement with platform benchmarks where available, but treat industry averages as directional rather than definitive. Your audience is not the average audience.
Step 4: Decide How You Will Handle Attribution
Video sits predominantly in the upper and middle funnel. Last-click attribution will systematically undervalue it. You need a position on this before the campaign launches, not after. Options include data-driven attribution in GA4, multi-touch models, media mix modelling for larger budgets, or simply being explicit about the limitations of your attribution model in how you report results.
Data-driven marketing requires making deliberate choices about attribution methodology, not defaulting to whatever the platform reports. The platforms will always show you a number that makes their channel look good. Your job is to triangulate across sources and apply judgement.
Step 5: Build Reporting That Earns Trust
The best video measurement report I have ever seen was three pages. It stated the objective, showed the KPIs against targets, explained what worked and what did not, and made a clear recommendation for what to do next. It did not lead with total views. It did not hide poor performance behind impressive-sounding secondary metrics. It was honest, and because it was honest, the client trusted it.
Tools like Sprout Social’s Tableau integration can help consolidate video data across platforms into cleaner reporting views, which reduces the manual reconciliation burden without removing the need for analytical judgement.
Platform-Specific Considerations
Each major platform has quirks in how it measures and reports video performance. Knowing these prevents misinterpretation.
YouTube
YouTube’s 30-second view threshold (or full video if shorter) is the most meaningful standard view definition in the industry. YouTube Analytics provides watch time data at a granular level, including audience retention curves that show exactly where viewers drop off. These retention curves are genuinely useful for creative diagnosis. If 60% of your audience drops at the 12-second mark, that is a specific, actionable signal. Custom GA4 reports can extend YouTube’s native analytics by connecting video engagement to site behaviour after the click.
Meta (Facebook and Instagram)
Meta’s three-second view threshold is low enough that autoplay in feed produces view counts that look impressive and mean relatively little. Focus on ThruPlays (views to completion or 15 seconds, whichever is shorter) and 10-second views as more meaningful engagement signals. Meta’s attribution window settings have a significant effect on reported conversions, so make sure you understand what window is applied before you interpret ROAS figures.
LinkedIn counts a view at two seconds with 50% of the video in view. For B2B campaigns where audience quality matters more than volume, focus on engagement rate, company-level reach (available in Campaign Manager), and lead gen form completions rather than raw view counts. LinkedIn’s audience data is more reliable than most platforms for B2B targeting, but the video metrics are not more reliable than anyone else’s.
Connected TV and Programmatic Video
CTV has become a significant channel for brands with meaningful video budgets. The measurement landscape here is still maturing. Completion rates on CTV are high because viewers are less able to skip, which can make the metric look impressive while telling you little about impact. Frequency capping is critical. Reach and brand lift studies are more reliable indicators of effectiveness than completion rate alone.
The Metrics Most Teams Overlook
Beyond the standard dashboard metrics, there are signals that often get missed but carry real diagnostic value.
Replays. If a meaningful percentage of viewers replay sections of your video, that is a strong signal of genuine engagement. It is also a signal that something in the content warranted a second look, which is worth understanding.
Share rate. Sharing is a high-intent action. It requires the viewer to associate their identity with your content. A video with a high share rate relative to views is doing something right at the creative level. Content marketing metrics frameworks often treat shares as a top-line engagement metric, but for video they are worth tracking separately from likes and comments because the intent is qualitatively different.
Post-view search behaviour. Brand search lift after a video campaign is one of the cleaner signals that awareness work is translating into intent. It requires search data alongside your video data, which most teams do not connect, but the correlation is worth looking for. Connecting ad activity to conversion tracking has been a theme in performance marketing for years, and the principle applies to video just as much as it does to search.
Audience retention at specific moments. Most platforms provide frame-level or second-level retention data. If you have a key message, product reveal, or call to action at a specific point in the video, you can see exactly how many viewers reached that point. This is far more useful than aggregate completion rate for understanding whether your message was actually delivered.
When I was growing iProspect from a team of 20 to over 100 people, one of the things I pushed consistently was building reporting that connected channel metrics to business outcomes rather than stopping at platform-level data. It was not always comfortable. Sometimes the numbers told clients things they did not want to hear. But it built a level of trust that surface-level reporting never could. The same principle applies to video measurement.
If you are working to build a more rigorous analytics practice across your marketing channels, the Marketing Analytics and GA4 hub covers the broader measurement landscape, from attribution modelling to custom reporting and beyond.
A Note on Testing
Video creative is expensive to produce and easy to assume is working based on partial data. Structured testing changes that. Running two versions of a video with different opening hooks, different calls to action, or different lengths gives you real data on what drives performance rather than opinions about what should drive performance.
A/B testing in GA4 can extend platform-level creative testing by connecting video variants to downstream site behaviour, giving you a fuller picture of which creative approach actually produces better business outcomes, not just better platform metrics.
The principle I applied early in my career still holds. When I taught myself to code rather than accept that a new website was impossible, the underlying logic was the same: do not accept the first answer if the problem is worth solving. Most video campaigns are under-tested because producing variants feels expensive. The cost of running on underperforming creative for three months is almost always higher.
About the Author
Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.
