IAS Flags Mobile Web as the MFA Hotspot. How Real Publishers Stay Off the List

Integral Ad Science published its 21st Media Quality Report on July 9, and one finding should get the attention of every publisher with a mostly mobile audience. According to IAS, mobile web display accounted for 45.1% of the open web impressions it analyzed in 2025, but 71.9% of made-for-advertising (MFA) impressions and 71.5% of ad clutter impressions. It also carried 54.9% of brand suitability failures.

The MFA rate itself tells the same story. IAS measured it at 2.0% on mobile web display versus 0.5% on desktop, four times higher. Its explanation is structural: mobile pages tend to have higher ad density and faster scrolling that shortens time in view, which gives low-quality inventory more room to spread.

For buyers, the takeaway is tighter quality controls on mobile web. For legitimate publishers, the risk is getting caught in those controls. Entertainment, celebrity and pop culture sites, which run long mobile sessions, galleries and a lot of ads per page, are exactly the kind of inventory a stricter filter can sweep up by mistake.

What IAS means by MFA and clutter

Definitions matter here, because they describe what the filters are looking for. In the full report, IAS describes MFA sites as properties built mainly to generate ad revenue rather than serve an audience, typically with low-quality or auto-generated content, high ad-to-content ratios and little editorial value.

Ad clutter is defined as environments where too many ads appear at the same time, lowering the chance that any single ad gets noticed. IAS found clutter rose from 0.3% of impressions in January 2025 to 0.8% in December, and called it one of only two major metrics that got worse over the year. The other was invalid traffic, which held around 1.1% globally, with the U.S. highest among major markets at 1.40%.

Notice what those definitions have in common. None of them requires a site to be fake. A real entertainment site with real writers can still show a high ad-to-content ratio on a phone, or stack three units in a single screen during a gallery. Classification systems see the page, not your intentions.

Why this costs real publishers money

When a buyer tightens MFA or clutter avoidance, the effect on the sell side is quiet. Bids do not arrive as errors; they just stop arriving. A campaign with a pre-bid MFA filter will not bid on pages it has classified that way, and exclusion lists built at the domain level can block an entire site because of a few templates. You see it later as falling bid density from certain DSPs, lower CPMs on mobile than on desktop for the same content, or deals that underdeliver without explanation.

The report also makes a positive case you can use. IAS measured video viewability at 79.7% in 2025, 11.8 points above display at 67.9%, and it ties viewability to the industry's move toward attention-based buying. Pages that give ads room to be seen are the ones that benefit when buyers pay for quality.

A mobile quality audit for entertainment sites

1. Measure your ad-to-content ratio on a phone

Load your ten highest-traffic templates on a real mid-range phone: article, gallery, list post, video page, homepage. Scroll at a normal pace and count how many screens show more ad than content. Do it logged out, with a clean browser, from a U.S. IP address, the way a classification crawler would. If any template regularly shows two display ads in one screen, fix that template before anything else.

2. Look hard at galleries and slideshows

Galleries are where entertainment sites most often resemble MFA patterns: thin text, one image per page view, an ad on every slide. If you run them, give each slide real editorial content, cap ads per gallery rather than per slide, and avoid designs that make a reader tap through dozens of pages to finish a single story.

3. Know where your traffic comes from

MFA is closely associated with buying cheap traffic and reselling it as ad impressions. If you buy traffic from social or native networks to hit audience targets, know which campaigns drive which sessions, and check whether those sessions look like your organic readers: time on page, pages per session, return visits. Traffic that exists only to be monetized is what the filters are designed to catch.

4. Ask how you are classified

You usually cannot see a buyer's exclusion list, but you can ask. SSP partners often have reports on why bids were filtered, and some will tell you when a domain is flagged by a verification vendor. Compare DSP-level bid rates on mobile and desktop for the same content. A large gap concentrated in a few buyers is worth a direct conversation.

5. Keep invalid traffic low and prove it

With IVT one of the two metrics that worsened in 2025, buyers will keep watching it. Run pre-bid IVT filtering, block known bad referrers, and keep a monthly record of your own IVT rate from an independent vendor. It is much easier to get off a list when you have the data ready.

6. Shift some mobile inventory to formats that earn more attention

Fewer, better placements usually beat more of them. A single well-placed video or large-format unit can outperform two stacked banners, and IAS's viewability data supports the case for video where it fits the content. Test removing your weakest mobile slot and watch total page revenue, not revenue per slot.

What to tell your sales team

Direct buyers and agencies are reading the same report. Get ahead of it with a short quality sheet: your mobile ads-per-screen policy, your IVT rate, your viewability by template and how you source traffic. For entertainment publishers competing with thin sites for the same celebrity and streaming searches, showing the difference is a selling point, not just a defense.

The IAS numbers do not mean mobile web is bad inventory. They mean it is where buyers will concentrate their filters. Make sure yours is the inventory those filters are meant to let through. If you work with a managed partner such as HBDR, a quality review of your mobile templates is a reasonable request to make this quarter.