The KPI Blind Spot: What Less Than 25% of CTV Advertisers Are Tracking
Most CTV performance reports are very good at answering one question and almost silent on another. They'll tell you exactly how many impressions have finished playing. They usually won't tell you how many of those impressions had a real person behind them.
Most advertisers aren't ignoring fraud on purpose. They're measuring what their reporting stack was built to measure, and most CTV dashboards were built around delivery, not authenticity. We touched on where fraud hides in our last post. This time we're going after the measurement gap that lets it stay hidden.
The metric everyone skips
When you ask most media buyers what they track on a CTV campaign, you'll hear the same handful of answers: completion rate, reach, frequency, maybe viewability. When asked if that traffic was verified as non-fraudulent, the room gets a lot quieter.
DoubleVerify's research found that less than a quarter of advertisers currently treat fraud or invalid traffic detection as a key performance indicator for their CTV campaigns. Not a footnote buried in an appendix. A KPI, tracked and reported right alongside everything else.
That's not because advertisers don't care. It's because most reporting stacks were built to answer, "did the ad run," not "did a real person see it." Those are two different questions, and only one of them shows up by default on most scorecards.
Curious where fraud or invalid traffic sits on your own KPI list? Ask us for a free fraud exposure check, no obligation, just a straight answer.
What a bundle ID actually tells you
A bundle ID is the piece of code that identifies which app an ad ran inside of. In theory, it's a simple way for a buyer to confirm they got what they paid for: real inventory in a real app, not cheap inventory dressed up to look premium.
In practice, it's one of the easiest things to fake in CTV. Pixalate found that on some devices, more than half of the app bundle IDs measured were malformed, unidentified, or outright fraudulent. Swap a bundle ID and cheap inventory reads as premium on a report that looks completely normal.
Add layered bot networks built to mimic human viewing behavior on top of that, and you get a campaign that hits its delivery numbers but a meaningful chunk of it was never watched by anyone. CTV commands some of the highest CPMs in digital advertising, and fraud follows the money every time.
Want to see what real bundle ID and device verification look like on an actual media plan? Let's walk through it together.
Questions worth asking before your next flight
You don't need a new vendor to start closing this gap. You need better questions for the ones you already have.
- Is fraud or invalid traffic tracked as a KPI on our CTV campaigns, or are we only looking at completion and reach?
- Do we know what a legitimate bundle ID looks like for the apps we're buying against?
- Is verification running for the entire length of the flight, or only checked once at kickoff?
- How many vendors sit between us and the real inventory, and what does each one actually see?
None of these questions require blind trust in the answer. A partner running real verification should be able to show you, not just tell you.
The bottom line
Fraud doesn't show up because advertisers are careless. It shows up because most CTV reporting was built to track delivery, not authenticity, and fraud is built specifically to exploit that blind spot.
Fraud-Free CTV treats fraud and invalid traffic as a KPI from the first bid, not an afterthought. Bundle IDs and device signals get verified continuously, not once at kickoff. And clients can see exactly what got filtered out and why. On average, that adds up to about 50 percent of the budget that fraud and inflated fees would have otherwise taken.
This same standard extends to Engage 4U, our unified platform for planning and buying CTV and digital together. Fraud analysis runs pre and post-bid as a built-in part of the platform, not an add-on, and clients using it have seen a 30%+ lift in working media as a result.
Read This One Alongside Last Week's Post
This post covers the measurement gap. Last week's post covers the fraud itself: spoofed apps, bot traffic, and why a clean report can still be hiding a dirty campaign. Together, they cover both what to look for and where to look for it.
Read Last Week's PostReady to see what your CTV report isn't showing you?
Let's talk through what fraud and invalid traffic tracking could look like on your next flight. No pressure, just a conversation about where your budget is actually going.
Talk to CTV MediaPrefer to see the numbers for yourself first? Request a free audit of your current CTV inventory.
- DoubleVerify, 2026 Global Insights Report: Must-CTV, Streaming's Shift From Promise to Performance (May 2026)
- Pixalate, Q1 2026 Malformed & Fraudulent CTV Bundle IDs Risk Report (May 2026)