A year ago, the Association of National Advertisers (ANA) (ANA) published the first part of the "Programmatic Supply Chain Transparency Study." However, it should have been called the "Programmatic Supply Chain Opacity Study."

The report and a follow-up study published in December revealed significant losses in programmatic advertising. $22 billion annually is spent in the USA on MFA sites, convoluted supply chains, fraud, and other manipulations.

On average, a programmatic ad campaign is shown on 44,000 sites, and more than 20% of these are MFA sites. In general, it’s a complete mess.

However, ANA and TAG TrustNet data show some improvements.

ANA vs. MFA

Last Monday, ANA and TAG TrustNet released the first report in a series of quarterly reports combining data obtained from various advertisers.

The goal of this study is to create a metric that helps advertisers understand how their campaigns comply with standards for combating MFA, media quality, CPM, stability, and DEI.

"2023 caused concern in the industry: 'Maybe you didn’t notice, but it seems like you have something out of control,'" said ANA CEO Bob Liodice. "This study helps advertisers ask the right questions and perform such checks regularly."

According to the analysis of data from 11 ANA member advertisers, the share of ad budgets spent on MFA has decreased from 15% to 4%.

ANA believes that at least part of this decrease is due to MFA attracting press attention after last year’s transparency reports were published.

As many advertisers seek to reduce the number of SSPs and partners in the supply chain, the average number of sites on which a programmatic ad campaign is shown has decreased from 44,000 to 23,000.

"These reports have helped our clients understand that transparency is important," said OMD CEO George Manas. "It's time to focus even more on this idea," especially now that video is increasingly diving into programmatic and there is a risk of repeating old mistakes.

(CTV ad campaigns were not included in this ANA study but will be included in the next one.)

No Peace

But despite the successes, there is still much work to be done.

Other companies present a less rosy picture. For example, according to a recent study by DoubleVerify, the number of paid impressions on MFA sites increased by 19% in 2023 compared to 2022.

Although 23,000 sites represent progress, there are still many sites that do not meet the standards of "premium internet" recently announced by The Trade Desk.

Tim Brown, head of Fiducia, a technology partner of TAG TrustNet, claims that the stated number of 23,000 is just an average. A detailed examination of the data reveals a more accurate picture.

Fiducia analyzed 400,000 sites and found that some advertisers only appear on 3,000 sites, while others only appear on 500. This means that advertisers choose different strategies: some care about efficiency, while others focus on reach.

"It's hard to believe that someone is trying to achieve a large reach simply by increasing the list of sites," says Brown. "But it's one of those things I constantly hear from advertisers: they need as many sites as possible for reach. But as I've said before, cheap can be expensive."

If your key performance indicator (KPI) is audience size but you don't see results, then it's not a very good KPI.

According to Manas, after the pandemic, many advertisers focused on short-term metrics such as clicks and views instead of long-term goals like brand awareness, customer loyalty, and LTV.

What’s Real?

This is why ANA supports what they call the TrueKPI framework — a set of truly meaningful indicators. The TrueKPI system, released by ANA in December, helps advertisers evaluate impressions based on their true value, not cost.

Consider an example: an advertiser pays $5 for 1,000 impressions (CPM), acquiring 10 million impressions, totaling $50,000 in advertising expenses. If the advertiser analyzes these impressions in terms of quality, they may find that only 7 million impressions were valuable, and the remaining 3 million were not. Thus, 3 million impressions were essentially wasted, with a CPM cost of $5, resulting in $15,000 in losses.

In this case, the true CPM would be $7.14. It’s not as cheap as it seemed.

Of course, an advertiser has the right to spend their ad dollars as they see fit. The main thing is that they understand what they are doing.

"This system helped us understand that the MFA problem is solvable; it helps evaluate quality and cost," says Liodice. "We should continue to work in this direction."

TrueKPI

Key aspects of TrueKPI include:

• Log-level data collection (LLD). TrueKPI uses log data from DSPs and SSPs to analyze each ad impression in detail. This allows for a precise evaluation of how each impression meets quality criteria.
• TrueImpressions — impressions that meet predefined quality requirements. Requirements can include the absence of fraud (IVT), visibility according to MRC standards, compliance with specific advertiser requirements, and exclusion of impressions on MFA sites.
• TrueCPM (true cost per thousand impressions) is calculated by dividing the total ad budget by the number of TrueImpressions multiplied by 1,000. This allows advertisers to compare standard CPM with TrueCPM and identify potential for optimization.
• myTrueImpressions customization. Advertisers can adapt the TrueImpressions metric to their needs by adding additional requirements, such as excluding impressions on blacklisted sites.
• Strategy optimization. With TrueKPI, advertisers can adapt their placement strategy, working to reduce TrueCPM by increasing the share of high-quality impressions in the overall volume of advertising. This may include revising suppliers, optimizing the ad supply chain, and more precise targeting.
• Reports and benchmarking. The system provides regular reports and comparisons with industry standards, allowing advertisers to evaluate the effectiveness of their campaigns compared to competitors and identify areas for improvement.

Other materials on this topic:

• Advertising visibility and safety in Russia in 2023
• Publishers earn from content farms
• New DeepSee product evaluates publisher quality
• Advertisers paid for ads on Forbes.com, but some ads ended up on a fake site