Daily Miracle 2.0: How Does Your Programmatic Technology Stack Up?

When it comes to programmatic advertising, publishers spend less than 20 percent of their time on value-related tasks like selling, upselling and lead generation. The remaining 80 percent of the time is spent on administrative activities such as billing, reconciliation, reporting, rekeying data, and monitoring inventory for ad fraud.

Now that programmatic trading is quickly moving from infancy to adolescence, it’s time for publishers to look at ways to optimize the entire workflow process. As reported in a 2015 Boston Consulting Group (BCG) study on programmatic profit for publishers, one of the keys is “assembling the right technology, both as an efficient way to access demand and as a decision engine to maximize revenue.”

The BCG report goes on to say that “Publishers must navigate a complex technology landscape. Assembling the right layered collection of software—or ‘stack’—is critical to revenue generation, since it is the stack that helps publishers maximize revenue by providing essential inventory management and process controls, as well as the gateway to advertiser demand.”

Single-Source or Point Solutions

To continue the child development metaphor, most publishers approach programmatic advertising in a manner similar to the way parents acquire clothing for a middle child. Buy some new things, repurpose old things, accept a few hand-me-downs, and maybe even create a couple of things in-house.

This approach seems almost necessary in a world where the advertising landscape changes so rapidly. As an example, the need for supply-side or sell-side platforms (SSPs) arose because publishers had to keep pace with advertisers and agencies that were already using demand-side platforms (DSPs) to make automated ad buying more efficient. SSPs were added to publishers’ advertising ecosystems with various levels of integration—both automated and manual—to existing ad servers, exchanges, ad management and billing systems.

The problems here are not dissimilar to other point solution models in different parts of the business. Each new product adds a level of complexity. New user interfaces need to be learned. Reporting, analytics and data interfaces need to be integrated with existing systems. Additional critical points of failure are introduced into the workflow. In addition, BCG notes that up to 10 percent of ad impressions can be lost during each stage of the programmatic workflow process.

As publishers look to optimize their programmatic technology stacks going forward, the single-vendor approach (or at least the as-few-vendors-as-possible model) appears to be the winning formula. This is especially true when it comes to ad servers and SSP ad exchanges. When provided by a single vendor, an ad server that’s fully integrated with the SSP will yield less than 0.5 percent lost ad impressions. Lost impressions mean lost revenue. So, publishers should make sure the integration points between the ad serving and ad exchange processes are as unified as possible.

Make Fraud Detection a Top Priority

Ad fraud is the dirty little secret in today’s world of programmatic trading. It’s causing lost revenue, lost credibility, and lots of manual intervention.

According to a September 2015 Bloomberg Business report, “Fake traffic has become a commodity. There’s malware for generating it and brokers who sell it. Some companies pay for it intentionally, some accidentally, and some prefer not to ask where their traffic comes from.”

Advertiser trust and customer credibility are the linchpins that distinguish news media publishers from lower-quality brands. Unfortunately, the publisher’s programmatic ecosystem is often ground zero for ad fraud, which degrades brand value and leads to skepticism, especially among local advertisers. Simply put, publishers need to take responsibility for ensuring that ads are served to real humans rather than bots or fraudulent eyeballs.

Every publisher’s programmatic technology stack must include tools to monitor inventory and guard against fraudulent impressions. Several software applications are available today to help publishers make certain their inventory is clean. Products from companies like Integral Ad Science, WhiteOpps, Pixalate, and DoubleVerify enable publishers to protect their web and mobile sites from becoming unintentional channels for fraud.

In a perfect world, automatic fraud detection will be part of every publisher’s unified SSP and ad serving system. Products like the AppNexus Publisher Suite is one example, combining ad serving, yield management, SSP, quality control, and fraud monitoring in an integrated platform. But, even a complete end-to-end solution is not always enough to prevent fraudulent traffic. Late last year, AppNexus openly acknowledged that transactions on its platform fell 65 percent after filtering its inventory for fraud.

It’s scary to note that a recent Association of National Advertisers (ANA) study found that for premium publishers—like news media companies—display ad buys with CPMs of $10 or higher had a 39 percent higher rate of fraudulent (non-human) views than average sites. For video ads with CPMs of $15 and above, the fraudulent bot rates were 173 percent higher than average.

Reporting on the ANA study for Publishing Executive, Scott McDonald of Nomos Research explains this phenomenon as a digital equivalent of the “Willie Sutton effect.” Just as Sutton famously said he robbed banks “because that’s where the money is,” fraudulent botnets are always looking for higher-value sites. News media publishers are prime targets for fraud because of the quality of their inventory and the higher CPMs they are able to command for premium ad buys.

The solution to the ad fraud problem lies in optimizing the programmatic technology stack. Similarly, to make the overall programmatic workflow more efficient, publishers must tip the scales in favor of value-generating activities over administrative chores.

Programmatic trading is maturing before our very eyes. To succeed and grow in the programmatic arena, publishers need to adopt a multi-pronged business model that includes: 1) fewer point solutions; 2) closer integration between various ad tech components; and 3) better data analytics to maximize yield, increase revenue, and improve overall customer satisfaction.

Peter Marsh is the VP Marketing, Newscycle Solutions. Daily Miracle 2.0 discusses the challenges, opportunities, trends and technologies impacting news media companies in the evolution from daily ink-on-paper products to round-the-clock digital offerings delivered in print, online, on tablets, and across mobile and emerging channels.