Editorial Team Founders trying to figure out if their CTV advertising is actually working got a sharp reality check this week. The big takeaway is that ad measurement for connected TV is still a complicated, murky problem that requires a smarter startup strategy if you want to see any real ROI. An industry report dropped at the Digital Marketing Summit in Austin showed a huge gap between what marketers *think* their ads are doing and what they can actually prove, which has a lot of founders wondering where all their CTV money is going.
Key Takeaways
- You have to build a unified measurement system. This means getting your CTV ad exposure data to talk to your first-party sales data (from Shopify, for example) so you can actually connect a view to a sale.
- Focus on incrementality testing instead of last-touch attribution. Use geo-experiments or holdout groups with a partner like Measured to isolate the real impact of your ads and see what sales you wouldn’t have gotten otherwise.
- Use server-side ad insertion (SSAI) from providers like Publica to stitch ads directly into the content stream, which gives you more accurate impression counts and cuts way down on ad fraud.
- Set clear, business-focused KPIs for CTV. Don’t just count views. Track metrics that matter, like website visits, app downloads, or even in-store traffic driven by your campaigns. A DTC brand that switched its focus to cost-per-site-visit completely changed its results.
The Measurement Conundrum in CTV
It’s easy to see why startups are drawn to CTV advertising. You get the huge audience of traditional TV combined with the specific targeting of digital ads, which looks like the perfect way to build a brand and get customers. The problem is that proving its direct effect on your revenue is a nightmare. Unlike digital ads where you can track every click, CTV is a fragmented mess of smart TVs, streaming sticks, and different content platforms. “The primary issue isn’t a lack of data, but a lack of cohesive, standardized data across platforms,” said Dr. Evelyn Reed, a top ad-tech analyst, in her keynote at the summit. “Each major streaming service, from Hulu to Roku, often provides its own set of metrics, making true cross-platform comparison and attribution a nightmare for lean startup teams.”
Because of this fragmentation, everyone falls back on proxy metrics like video completion rates (VCR) or total impressions. While those numbers tell you if people are watching, they don’t tell you if they’re buying. I talk to frustrated founders all the time who are pouring cash into CTV campaigns with no clear line of sight to ROI. For example, a small e-commerce startup might get great completion rates on their ads but can’t find any corresponding lift in their analytics. The attribution just breaks. It completely misses the common scenario where a customer sees an ad on their TV Tuesday night and then finally makes a purchase on their phone Friday afternoon. Your standard last-click or multi-touch models just can’t track that journey, making the whole campaign look like a failure when it might not be.
| Measurement Strategy | Traditional Last-Touch Attribution | Proxy Metrics (e.g., VCR) | Unified Measurement & Incrementality Testing |
|---|---|---|---|
| Directly links to sales/conversions | ✗ Often falls short | ✗ Does not directly translate | ✓ Yes, with first-party data |
| Captures full customer journey | ✗ Skews perceived performance | ✗ Limited insight | ✓ Addresses multi-device path |
| Requires cohesive, standardized data | ✗ Struggles with fragmentation | ✗ Relies on platform data | ✓ Integrates diverse data sources |
| Isolates true ad impact | ✗ Difficult to isolate | ✗ Indirect correlation | ✓ Uses geo-experiments/holdouts |
| Leverages first-party data | ✗ Typically relies on ad platform | ✗ Not a primary focus | ✓ Achieved 15% conversion increase |
| Addresses ad fraud/impression accuracy | ✗ Limited ability | ✗ Not directly addressed | ✓ Via SSAI technology |
| Suitable for lean startup teams | ✓ Easier to implement initially | ✓ Simple to track | ✗ Requires upfront investment/planning |
Implications for Startup Strategy
For a startup where every dollar has to count, this measurement problem is a critical strategic impediment. Founders need to plan for measurement before they ever buy a single impression. This means going way beyond the basic reports from your ad server. You have to prioritize integrating your CTV ad exposure data with your own first-party customer data, likely using a customer data platform (CDP) to act as the glue. This is how you connect an anonymous ad view on a smart TV to a known customer in your database. A Reuters report from earlier this year confirmed this works, finding that businesses that did this saw a 15% average jump in attributable conversions over those just using platform metrics.
This also means getting serious about incrementality testing. You need to stop asking “Did CTV ads drive sales?” and start asking “Did CTV ads drive *additional* sales that wouldn’t have happened anyway?” To do this, you have to run controlled experiments, like targeting one city with ads while leaving a similar city as a control group. The difference in sales or other KPIs between the two groups is your true, incremental lift. This is where a startup can actually beat bigger competitors. Large companies are often too slow and bureaucratic to run these kinds of tests, but founders can build this methodology into their marketing DNA from day one and run circles around them.
What’s Next for CTV Ad Measurement
The industry is trying to clean up this mess with more standardized tools. People are working on open-source measurement protocols and cross-platform identity graphs (like UID2 from The Trade Desk) that promise to fix some of the fragmentation. Major players like The Trade Desk and Magnite are sinking money into building more unified reporting. But let’s be realistic, these solutions are going to take a while to become the standard. So what do you do right now? You have to be proactive. Demand more transparency from your ad partners. Ask for the raw impression data and detailed audience reports. What specific data points can you get? Can you verify them? You may even need to invest in an in-house data analyst. It’s an expense that will almost certainly pay for itself by preventing wasted ad spend.
Founders need to see CTV ad measurement as a core business strategy, not some technical task for the marketing team. By integrating their own data, focusing on incrementality, and demanding real answers from partners, they can turn CTV from a money pit into a growth engine. The well-known problems with CTV ad transparency are a perfect example. They’re a huge hurdle because they obscure where your money is actually going. Getting a grip on this requires a strong startup governance mindset to manage risk. Without clear measurement, even the most advanced CTV ad personalization efforts will fail.
Why is CTV ad measurement more complex than traditional digital ad measurement?
It’s complex because the CTV world is fragmented across countless devices and streaming platforms. Each one reports data differently, so there’s no single source of truth like you get with web-based ads that have simple click tracking.
What is incrementality testing in the context of CTV advertising?
It’s a method for measuring the *additional* sales or conversions you got directly from a CTV ad campaign, the ones that wouldn’t have happened otherwise. You typically do this with controlled tests, like showing ads in one geographic area but not a similar one, then comparing the results.
How can first-party data improve CTV ad ROI measurement for startups?
By connecting your own customer data (like from your website or sales platform) with CTV ad exposure data, you can finally link an anonymous ad view on a TV to a specific person buying something on your site. This gives you a much clearer path to attribution and shows you how ads impact real customers.
What are some key KPIs startups should focus on for CTV campaigns beyond impressions?
Forget just tracking impressions or how much of the video was watched. You need to track KPIs that actually affect your bottom line: website visits, app downloads, lead form submissions, and, most importantly, direct sales or sign-ups that can be tied back to the ad campaign.
What role does server-side ad insertion (SSAI) play in CTV ad measurement?
SSAI gives you a much more accurate count of ad impressions. It works by stitching the ad directly into the video stream on the server side, which helps bypass ad blockers and gives you more reliable data on whether an ad was actually delivered and seen.
