• by Amanda Bleich , Op-Ed Contributor, August 10, 2017

To say that programmatic has grown in popularity is an understatement. Close to 80% of all digital media buying will be done programmatically in 2017, according to eMarketer. With this in mind, advertisers no longer have to worry about negotiating hundreds -- or thousands -- of direct media buys, and can reach audiences at scale while using efficient pricing and advanced algorithms.

Still, as with all great things, there’s still room for improvement. There are plenty of ways programmatic advertising can go wrong. Anyone with a few years’ experience in this industry has likely seen this happen.

While that’s no reason to avoid programmatic entirely, perhaps it’s time to proceed with caution. Consider a new approach: Advertisers take responsibility to be well-versed on the potential problematic areas, then arm themselves with actionable plans of what to do when things go south.

One of the first problems many programmatic practitioners run into is related to data. There are iindustrywide issues with how data providers are vetted, how data is gathered and whether it is verified. Many marketers pay premium CPM rates with the idea that they’ll reach a precise consumer, but instead, ads for dog food hit people who don’t own any pets, and ads in Spanish appear in front of consumers who only speak English.

There’s also an increasing overreiiance on technology: the misguided “set it and forget it” mentality. While demand-side platforms (DSPs) offer sophisticated solutions that automate and streamline the process, it's almost impossible to reach desirable KPIs without human guidance. No DSP will simply turn on and produce a perfectly implemented and executed campaign automatically. The technology needs to be monitored and calibrated over time to provide maximum efficiency. It requires human participation to monitor results, to decide on necessary campaign adjustments -- and then to implement optimizations.

Finally, there are the headline-worthy matters that keep CMOs up at night. Bots, fraud and poor viewability are enough to scare the most conservative brands away from investing in programmatic. No one wants to buy inventory that never reaches actual consumers, either because the ad call came from a bot or because the ad loaded so far down the page that no one sees it.

Then there’s transparency, which has been an advertising buzzword for so long that we wonder how it can even be up for debate. If everyone requires transparency regarding publishers, fees, and even platforms used, then why are marketers still forced to ask for it?

Rectifying these mistakes, and preventing them from happening in the next campaign, ultimately comes down to education and direct insistence for what you and your client want, no matter where your company falls in the programmatic landscape.

Self-education is a major first step. It’s impossible to emphasize how important it is for everyone involved in programmatic to take initiative and read up on the latest trends. Traders working within agencies need to be in contact with different data providers and DSPs to understand the latest innovations and updates. The DSPs themselves need to be as communicative as possible with every new release, showing clients what changed and what it means for future campaigns.

Many mistakes with programmatic arise out of a belief that every DSP, omnichannel or not, is a one-size solution that fits advertisers across verticals and with varying KPIs. Sadly, that’s not the case. Advertisers need a clear idea of what they want to accomplish, and then they need to ask individual DSPs and trading desks how their solutions will help achieve this goal. Advertisers may find themselves using multiple DSPs with differing functionalities best suited for specific campaigns.

At the end of the day, everyone in the programmatic ecosystem needs to know what’s out there and explore it accordingly. If you can anticipate the ways programmatic goes wrong, then you can safeguard campaigns against these issues from the very start.