There is a debate in the ad world right now about the danger of “made for advertising” websites but it seems that not everyone is clear on what constitutes an MFA—there is even disagreement over whether they are all that bad for brands.

The Association of National Advertisers has been shining a light on “made for advertising” sites, or MFAs, issuing a study in June claiming that at least $13 billion of the $88 billion programmatic ad market was going to what it deemed spammy, clickbait or low-quality sites. In other words, MFAs siphoned money from what the ANA would consider more worthy publishers, those more deserving of top brand dollars.

The trade group’s report, peaking into darker corners of the web, set off alarm bells among ad agencies and brands. Parts of the industry, including third-party viewability firms and supply-side ad platforms, began hunting for these thousands of “made for advertising” sites, promising to help cut them from the supply chain.

There are still a few wrinkles to iron out, however, as different stakeholders have varying takes on what is, and what is not, an MFA. Ad Fontes Media, for instance, manually rates websites in search of the spammiest ones, to help advertisers block them from programmatic ad buys. “MFAs monetize, strongly, some of the worst content on the internet,” said Vanessa Otero, founder and CEO of Ad Fontes Media.

More: Yahoo free of made-for-advertising inventory, audit finds

There is some nuance in this debate, because even respectable publishers, especially in local news categories, can get caught up in the recently revved-up MFA detection systems. Also, some digital ad buyers might still be interested in delivering ads all over the web in the cheapest way possible, and sometimes MFAs might actually work.

“We are finding that for some advertisers, avoiding this inventory is not necessarily a priority,” said Nicole Perrin, VP of business intelligence at Advertiser Perceptions, a research consultancy group.

With all these competing forces, Ad Age wanted to get to the bottom of MFAs, see how bad they really are, and understand why the problem needs to be addressed.

What is an MFA?

Most internet users probably don’t know what technically constitutes an MFA, but people stumble on them all the time, intrigued by a clickbait headline in a search result, a social media post or on another website. Often, headlines are about celebrities under the guise of, “you won’t believe what they look like now.” They can also be about lifestyle, such as “check out the world’s biggest yachts.”

“All of a sudden there’s eight ads within one frame for some of the biggest brands,” Otero said, “just monetizing that site.”

Ad load is one of the key definers of an MFA, according to new standards outlined by ANA just last week. ANA reached an agreement with other trade groups, including 4A's, the World Federation of Advertisers and the Incorporated Society of British Advertisers, about what they consider to be an MFA. ANA was semi-nebulous in its definition, giving some leeway, but it said that MFAs have high ad-to-content ratios, “usually at least twice the internet average.”

The trade groups’ standards also listed factors such as “numerous refreshing banner ads,” “autoplay video ads flood the site” and slide shows with multiple ads. Slide shows have long been a problem online, used as a tactic by publishers to get viewers to click through their web pages to get through the slides, serve ads and enlarge page view counts.

MFAs also appear to use an infinitely scrolling website, with ads that run throughout, as noted by Ad Fontes and others. 

DoubleVerify, a third-party measurement and viewability firm, has its own definition of MFAs, which includes high ad density and buying traffic. “Some sites might have a high ad density but, they’re getting direct traffic sources, and it’s unique content,” said Jack Smith, chief product officer, DoubleVerify. 

“Direct traffic” is one of the signs that a site might not be an MFA, because the reader is choosing to go to the site, not lured there by a clickbaity ad.

What are the risks with MFAs?

The interest in MFAs came after ANA issued the first part of its “Programmatic Media Supply Chain Transparency Study” at the Cannes Lions International Festival of Creativity in June. That’s where the trade organization first delivered the message to ad tech companies and digital ad buyers that ad money was possibly being wasted on these seemingly low-quality sites. The ANA found in the study that the average participating advertiser ran ads on 44,000 websites, and suggested that was an unmanageable number, likely putting brands into unsuitable settings. “A startling 15 percent of spending and 21 percent of impressions in this study appeared on sites made only for advertising,” the ANA wrote.

The ANA claimed that as it studied the issue further it could uncover $20 billion in wasteful ad spending in open programmatic markets, or 23% of the $88 billion industry.

The ANA identified several concerns with the state of the programmatic industry’s MFA problem: high-quality publishers lose a percentage of that revenue, defunding serious journalism; also clickbait sites could take away from minority-owned and operated media outlets, according to the report.

“When advertisers prioritize cost over value, they do so to their own detriment,” the report said.

Are there any benefits for brands in MFAs?

“Based on what I’ve seen, there are lots of advertisers who know that sometimes this inventory may be useful for them for reaching KPIs that they’re being judged on,” Perrin, the ad research consultant, said.

The content on MFAs can be attractive, too, because it’s innocuous, covering celebrities, health, lifestyle, fashion and less hard news. This is a double-edged sword though, according to critics of MFAs. Purveyors of such content intentionally create articles that will seem “brand safe” when opening their ad inventory on automated, internet ad markets. The MFAs avoid using keywords in the articles and headlines, which could set off blocklist alarms created by brands that don’t want to appear alongside certain words and phrases. Meanwhile, more prestigious publishers lose ad dollars when brands use “keyword” blockers that skip ad inventory within articles about polarizing subjects.

Also, many MFAs have human traffic, meaning bot-generated fraud is not necessarily a problem, according to DoubleVerify's Smith. In September, DoubleVerify launched a new service to monitor whether campaigns ran on MFAs.

“Clients are looking at the sites that are being classified as MFAs,” Smith said, “and determining how they want to deploy, if they want to deploy, and how much they want to deploy on those kinds of sites.”

“Most sophisticated advertisers are not just mass blocking,” Smith said.

How does generative AI factor into MFAs?

Generative AI is making it easier to create websites and populate them with fake content. The MFAs can also use generative AI to get around filters that might otherwise identify them as MFAs to the programmatic ad buyers.

One tell-tale sign of an MFA is that it contains articles copied from other websites. An MFA could run a network of sites all with the same stories, but AI-generated text, with tools such as ChatGPT widely available, make it easy to tweak the articles slightly. Also, legitimate publishers are using AI to create stories, too, so blunt-force blocking ads on AI-generated articles could ding those media companies, Smith said.

MFAs also use AI to evade keyword blocklists, as the creators of articles can program text-generators to avoid using the terms that brands ban, according to Otero of Ad Fontes. “A lot of that work is being outsourced to generative AI now,” Otero said.