How To Overcome Bad AI In Your Programmatic Marketing Efforts


The recent prominence of AI tools such as ChatGPT has raised questions about their utility across industries in the long term. Certainly, myriad applications have sprouted, creating everything from grocery lists to undergraduate term papers. But as major publishers eye its utility, marketers are growing concerned about how a resulting surge of AI-generated content might impact the advertising marketplace.

An influx of AI-generated content will significantly alter the programmatic marketplace, and as marketers, we must contend with a potential waste in ad spend against poor quality content amid inflated inventory. To future-proof our programmatic marketing efforts, the approach is two-pronged: We must address both the content and the audience to ensure profitability for our clients and their brands.

Contending With Content

As marketers navigate an AI-informed marketplace, we may actually be contending with a familiar foe: ad fraud. For the uninitiated, ad fraud in programmatic typically means fraudsters falsify impressions and clicks on programmatic campaigns by deploying bad practices like malware, bots or human fraud farms. As a result, advertisers lose money.

A recent study found that programmatic ad fraud, which totaled $65 billion in 2021, was more than double the size of credit card fraud that same year. And no wonder, as programmatic is big business—that same year saw $418 billion in global programmatic ad spend. As the programmatic business continues to grow, it can pose a tantalizing landscape for fraudsters.

The first line of defense marketers should deploy is careful consideration of the content they are advertising against. AI has enabled the rapid creation of made-for-advertising sites (MFAs), websites that prioritize a high volume of clicks but only deliver truly low-value impressions that are not productive for consumers or advertisers. These sites are often characterized by “low-quality content, such as fake news, conspiracy theories, or spammy links, and may use tactics such as pop-up ads, auto-play videos, or intrusive ads to maximize ad revenue for the site owner,” according to the Association of National Advertisers (ANA).

By carefully considering from whom we are purchasing, prioritizing private marketplace deals, and limiting the overall volume of publishers engaged, it becomes far easier to control for known entities and thus avoids inadvertently spending on an MFA. While this considerably decreases the number of websites on which we advertise, it also significantly lowers risk.

Addressing Audience

AI has supercharged audience fraud in programmatic. What had previously been plagued by human-staffed click farms is now combatting a super-charged bot version of this practice to falsify impressions and thus steal ad dollars. This has grown so rampant that a recent report from the ANA indicated that up to 23% of programmatic spend may be wasted, in no small part because of these fraudulent practices. Major players in the space have been developing supply path optimizations to address this issue, working to differentiate premium inventory from relevant publishers.

In addition to working closely with such partners, a marketer’s best defense against audience fraud is to partner closely with measurement teams to train buyers on a new type of evaluation process ahead of new buys. Measurement teams can help buyers vet inventory diligently, and with healthy skepticism, by looking past impression numbers and instead searching for irregularities in the data to alert for potential fraud. Retraining ourselves to consider impression farming to be as big of a concern as click farming will serve marketers well as fraud practices continue to evolve at a rapid rate.

All of this is to say that leaps in marketing technology are not to be feared, shunned or avoided. If we shrug off exciting developments, such as generative AI, our professional expertise suffers, as does the efficiency of our campaigns. Instead, we must learn as quickly as those who wish to adopt new tools for malevolent means in order to serve our brands and our industry.

This article first appeared on

Seeking to build and grow your brand using the force of consumer insight, strategic foresight, creative disruption and technology prowess? Talk to us at +971 50 6254340 or or visit

About Author

Comments are closed.