The CTV Paradox: Record Spend, Record Fraud
Connected TV has become the channel everyone wants a share of. Programmatic CTV spend reached roughly $36 billion globally in 2026, up 28% year over year, and now accounts for about 71% of all CTV advertising. The budgets are real, the audience is real, and the format works.
The problem is that fraud operators read the same forecasts. As money moved into streaming, so did the people looking to take a cut of it. This article looks at what the 2026 fraud data shows, why CTV is a specific target, and what advertisers can actually do about it.
What the data shows
DoubleVerify's 2026 Global Insights report on streaming, Must-CTV, gives the clearest recent picture. Its headline finding: the company detected 140% more CTV fraud schemes and variants in Q1 2026 than in Q1 2025. Fraudulent CTV apps grew tenfold in 2025 compared with 2024.
The cost estimate is equally direct. In campaigns running without verification, DoubleVerify puts the loss at around $1.8 million for every billion CTV impressions served. Given that trillions of impressions run each year, the total adds up fast.
Other measurements point the same way. Pixalate found that more than $1.1 billion in open programmatic CTV spend was lost to invalid traffic in a single quarter. Some industry estimates place total global CTV fraud losses near $7.4 billion.
Why CTV specifically
CTV fraud is not simply display fraud with a different label. A few structural features make streaming unusually attractive to bad actors.
The payout is bigger. CTV commands far higher CPMs than display. A fake impression that passes as CTV inventory is worth several times what the same effort would earn in a banner environment, so the return on building a fraud operation is much higher.
There is no browser to check. Display fraud detection leans on signals a browser provides. CTV runs through apps, devices and server-side ad insertion, which strips out many of those signals and makes a fabricated impression harder to distinguish from a real one.
The chain is long. Most programmatic CTV inventory passes through several intermediaries before reaching an advertiser. Every extra hop is another place where inventory can be misrepresented.
Measurement is still uneven. CTV standards are not yet consistent across devices and platforms. Without a common baseline, isolating fraud signals is harder than it should be.
The direct-deal assumption
One of the more useful findings in the DoubleVerify research is a correction rather than a statistic. Many buyers assume that buying CTV through direct deals or private marketplaces removes the fraud risk. The data does not support that. Gilit Saporta, who leads DV's Fraud Lab, put it plainly: "fraud always finds a way".
That matters because a lot of CTV budget has moved into deals specifically on the assumption that a deal is inherently clean. A deal narrows who can bid. It does not, on its own, verify what is being sold.
The measurable difference comes from verification, not deal type. In DV-protected CTV campaigns, fraud rates ran below 1%. In unprotected campaigns, the figure was close to 9%.
What to do about it
The gap between 9% and under 1% is the whole argument for treating this as an operational problem rather than an unavoidable cost.
Make invalid traffic a KPI. DoubleVerify found that fewer than a quarter of advertisers — 21% — measure CTV performance using invalid traffic or fraud detection as a KPI. If it is not on the report, nobody optimises against it.
Demand app-level transparency. Knowing which app an ad actually ran in is the single most useful piece of CTV reporting. The good news is that this is improving: CTV app transparency rose 16% in Q1 2026 compared with Q1 2025. Ask for it and check it.
Run pre-bid filtering, not just post-campaign reporting. Catching fraud after the fact recovers a report, not a budget.
Audit the supply path. Fewer intermediaries means fewer opportunities for inventory to be misrepresented along the way.
Do not treat a deal as protection. Apply the same verification to PMP and direct CTV inventory as to open auction buys.
The takeaway
None of this is an argument against CTV. The channel delivers attention and reach that few others match, and the spend growth reflects genuine performance rather than hype.
The paradox is simply that a channel growing this quickly attracts investment and exploitation at the same time. What separates advertisers who lose money in CTV from those who do not is rarely the size of the budget or the quality of the media plan. It is whether verification was treated as a line item from the start or as something to look at once results come in disappointing.
For teams planning CTV spend, the practical first step is small: pull the last quarter's CTV reporting and check whether app-level detail and an invalid traffic figure are in it. If either is missing, that is the gap worth closing before the next budget goes out.
