Key takeaways
- Direct and premium deals can still carry bot fraud, so the deal label alone isn't proof of quality.
- A deal sold as direct may run through networks and resellers, so know who is behind the supply.
- Verification, deal records and contract terms should cover every buy, not just the open exchange.
Many brands focus their verification and fraud prevention on open exchange buys and treat direct deals with premium publishers as safe. A September 2026 report from DoubleVerify's Fraud Lab challenges that habit. It describes several campaigns, bought as direct or guaranteed deals, that still delivered large volumes of bot traffic.
What the Fraud Lab found
In one case, a large retail brand ran a one-to-one direct audio deal with no exchange in the path. DoubleVerify says the campaign was hit with 20 million bot fraud incidents, nearly 6 million of them on a single programmatic guaranteed placement. It tied the activity to a botnet it calls BeatSting, which spoofs residential IP addresses and audio apps.
In another, a major financial brand used what it believed were six direct deals around a World Cup series. Of slightly over 1.2 million impressions, about 1 million were identified as sophisticated invalid traffic. The "publisher" was in fact acting as a network, reselling inventory from many smaller publishers with multiple SSPs in the path. A third case, a pharma brand's direct-to-publisher CTV package, delivered more than 700,000 impressions tied to the same bot scheme.
Why premium buys can get the least scrutiny
DoubleVerify notes that direct CTV deals generally carry a lower risk of sophisticated invalid traffic, and premium publishers remain a valuable part of most plans. The issue is the assumption that can follow. When a deal is positioned as premium, buyers may trust the placement and check the inventory less.
Lower risk isn't zero risk. Direct deals often carry the highest prices in a plan, so when they get the least scrutiny, the most expensive impressions are the ones going unchecked. As DoubleVerify puts it, "Premium is a commercial tier, not a security protocol."
Four things to consider
1. Check your coverage
Confirm that verification and fraud filtering apply to direct, programmatic guaranteed and private marketplace deals, not just open exchange buys. Measurement should continue after the buy across premium inventory, rather than treating the deal type as proof of quality.
2. Know who's behind the deal
Ask whether the seller is the publisher itself or a network reselling others' inventory. A deal labeled direct can still involve intermediaries, so the supply path should be clear before the money moves.
3. Keep records that travel
Make sure deal IDs and supply path details are captured consistently in reporting. When something looks wrong, those records are what let your team trace the problem to its source.
4. Write terms with teeth
Agreements with publishers and partners should spell out what happens when invalid traffic is found, including make-goods or refunds. Clear terms turn a verification report into a remedy.
Trust the partner, verify the buy
Premium publishers earn trust, and direct relationships are worth building. Every buy still needs verification, and the spend that costs the most deserves at least the same checks as the rest of the plan. We would start by mapping where verification stops today.
Questions to ask your team
Which of our deals, if any, run without fraud verification today?
For each direct deal, who is actually selling the inventory?
Are deal IDs and supply path details captured in our reporting?
What do our contracts say about refunds for invalid traffic?
From Insight to Impact
Want verification that follows the spend?
Our Strategic Media Brief and Channel Assessment review how your media is bought, where verification applies and what your partner terms cover, so every buy is held to the same standard. See what's included, and the sample pages behind them.
Sources
DoubleVerify Fraud Lab: Just Because It's a Direct Deal Doesn't Always Mean It's Fraud-Free (Sept. 18, 2026)Case study figures are DoubleVerify's own findings; the brands involved are not named.
