Mobile web display is now the top digital media quality risk, IAS finds
IAS finds mobile web display drives the largest share of suitability violations, clutter, and MFA impressions. What it changes for verification
Integral Ad Science (IAS) says mobile web display has become the biggest source of media quality risk, even as overall media quality improved across 2025. The company detailed the findings in its annual Media Quality Report.
Mobile web display now sits at the uncomfortable intersection of scale and fragility: it represents a large share of open web impressions, yet it concentrates several of the failures marketers are trying to reduce, from brand suitability issues to made-for-advertising inventory.
One strategic tension is hard to ignore. The industry has learned to optimize for performance signals, but many of the worst quality outcomes are still shaped by environment design, not targeting sophistication.
Table of contents
Jump to each section:
- Why mobile web display concentrates quality risk
- What the 2025 metrics reveal about waste and verification
- How AI changes fraud, MFA, and brand suitability
- Why video environments are winning on attention signals
- What this means for marketers
Why mobile web display concentrates quality risk
IAS found that mobile web display made up 45.1% of all open web impressions in 2025, but generated an outsized share of key issues. It accounted for 54.9% of brand suitability violations, 71.5% of global ad clutter, and 71.9% of all made-for-advertising (MFA) impressions.
A memorable way to frame it: when an environment is optimized for scrolling speed and ad density, quality problems do not show up as edge cases, they show up as averages.
IAS also pointed to structural characteristics that make the format more exposed, including higher ad density and faster scrolling behavior. In practice, that means viewability and attention are not just creative problems, they are page and feed mechanics problems.
One more datapoint sharpens the risk profile: IAS noted the MFA rate on mobile web display is now four times higher than on desktop. That gap suggests marketers cannot assume “mobile = desktop, just smaller.” The inventory dynamics are different.

What the 2025 metrics reveal about waste and verification
Even with broader improvements in media quality during 2025, IAS highlighted two metrics moving the wrong direction.
First, invalid traffic (IVT) remained broadly stable at 1.1%, underscoring that fraud remains persistent rather than episodic. The shift is less about whether fraud exists and more about how convincingly it can masquerade as legitimate supply.
Second, ad clutter rose from 0.3% in January to 0.8% in December, which IAS linked to higher ad loads during the holiday season. That is a reminder that “peak performance periods” can also be “peak degradation periods,” especially when supply paths get crowded and the incentive to monetize intensifies.
A useful observation here is that clutter is not merely a user experience issue. It is an auction-shaping variable. When density rises, the probability of your impression being both seen and contextually acceptable often drops, even if the buy looks efficient on CPM.
How AI changes fraud, MFA, and brand suitability
IAS noted that fraudsters are increasingly using AI to create fraudulent inventory that resembles legitimate publisher environments, with the threat expanding into connected TV as well.
That matters because the classic mental model of fraud, such as obvious bots and crude spoofing, is becoming outdated. The more interesting question is what happens when synthetic inventory looks “brand safe” at a glance but is still economically engineered to extract spend rather than deliver outcomes.
A second tension emerges: marketers want automation to manage complexity, but automation also lowers the barrier for adversaries to scale deception.
IAS’ view implies that brand suitability and MFA controls will need to be more granular, especially in environments where the combination of speed, density, and low-friction publishing makes it easier for low-quality pages to blend in.
Why video environments are winning on attention signals
IAS reported that video continued to outperform display on core delivery metrics. Global video viewability reached 79.7%, compared with 67.9% for display, an 11.8 percentage-point advantage.
IAS also found performance-linked outcomes for attention-based strategies in video-forward environments: 56% higher attention scores and a 76% reduction in cost-per-click. It also reported that high-attention video ads generated household-level sales lifts up to 313% higher than lower-attention placements.
A concise takeaway: attention is becoming a quality currency, not a creative flourish.
The implication is not that display is “bad,” but that the environments where people intentionally watch, pause, or lean in are structurally friendlier to verified outcomes than environments designed for rapid browsing.
What this means for marketers
If mobile web display carries a disproportionate share of suitability violations, clutter, and MFA, marketers will need to treat it as a distinct risk surface, not a default extension of cross-channel reach.
- Separate “reach scale” from “quality scale” in planning
Mobile web display can deliver volume, but the IAS ratios show that volume can come bundled with higher exposure to clutter and MFA. Build plans where incremental reach must clear explicit quality thresholds. - Use seasonality as a risk signal, not just a budget signal
IAS’ clutter increase through the year suggests that Q4 can change the effective media environment. Treat peak periods as times to tighten verification and suitability parameters, not loosen them. - Update fraud assumptions for AI-shaped inventory
If AI helps fraudulent inventory resemble legitimate environments, then simple heuristics and surface-level signals will be less reliable. The value shifts toward verification approaches that can detect patterns at scale. - Treat attention metrics as a control layer, not a reporting layer
IAS’ attention-linked outcomes imply attention can be used to shape buying decisions, not just to evaluate creatives afterward. That is especially relevant when the same spend can land in either high-density scroll environments or more engaged video contexts. - Make regional and vertical variance part of governance
IAS noted that risks vary by region and industry, with APAC showing lower MFA rates than North America and EMEA but higher-than-global-average suitability fail rates, and travel and entertainment seeing high IVT and MFA. One global standard may be too blunt.
The bigger shift is that “quality” is no longer a defensive requirement bolted onto media buying. It is becoming part of the performance stack.
As AI accelerates both content creation and deception, the winning discipline will be the ability to prove that an impression was not just delivered, but delivered in an environment that can plausibly drive business outcomes.
In that world, the best media teams will not argue about whether to buy scale. They will argue about which forms of scale are still worth trusting.

