Programmatic pitfalls: Four Mistakes to Avoid When Deploying Digital

Programmatic advertising remains the backbone of many digital advertising campaigns. As a result of technology advancements, programmatic is now more automated, more transparent, and more accessible than ever before.Across the digital advertising ecosystem, advertisers, media planners, and ad buyers continue to rely on programmatic to stay competitive and drive performance.But despite all the bells and whistles programmatic can provide, it is imperative to not just "set and forget" campaigns.These four common mistakes keep programmatic campaigns from reaching their full potential and — ultimately — hamstring overall success.1. Too many variables, and too little timeThere's a lot of targeting, creative, and inventory available with programmatic campaigns. By testing hypotheses around audiences, creative format and messaging, advertisers can find out what works. However, testing can quickly become underfunded or inconclusive if effective controls aren't set up during the building phase.Instead of throwing everything against the wall and seeing what sticks, advertisers should employ a test and learn strategy. Identity one or two variables to test.  Then allocate sufficient campaign budget and flighting to ensure there’s enough money and time to run proper testing.This involves setting performance expectations with clients during the learning phase. Upper funnel learnings can be invaluable to further inform campaign optimizations — think of it as a digital focus group — however, bottom funnel performance can be delayed as the campaign works through these learnings to set up efficiencies and effectiveness down the line.2. Being too specific with targetingTo make sure every dollar is spent efficiently, it's tempting to layer on a bunch of targeting options. Advertisers might not realize, but low scale, multiple data providers, and Boolean logic (the selection of "and" or "or" between segments) are the top three ways to drive up CPMs.Instead, advertisers should always look to use the same data providers, and default "and/or" equations to “or” whenever possible. In smaller geotargets (zip codes, cities, small DMAs), they should also keep an eye on formats that can be difficult to scale.3. A singular KPI across the funnelEven though programmatic advertising addresses all aspects of the funnel, it's important to know where each tactic is most effective. Every advertiser wants strong metrics in terms of ROAS, ROI, CPA, but unfortunately not all strategies and channels are structured to hit these objectives.KPI objectives should be applied where most appropriate. For example, holding video or audio to a hard cost-per-acquisition KPI, is unwise because that isn’t what these channels and associated creative messages are designed to do. Rather holding the entire campaign accountable to the same bottom funnel KPI, advertisers should differentiate their strategies accordingly to better plan, optimize and drive success.4. Optimizing without a scheduleIt is common for advertisers to spend a considerable amount of time determining what KPIs and benchmarks should be used, but rarely do they outline what types of optimizations will be made and when they will be made. And not all optimizations will be made with the same cadence.Targeting optimizations are best adjusted weekly at the micro level, and bi-weekly at the macro level. Creative adjustments can be effective as early as 3-weeks but more-often-than-not are best applied every 6-8 weeks.By scheduling optimizations that allow time for data-driven learnings marketers can apply meaningful insights and drive performance. Additionally aligning campaign testing periods, to campaign performance goals ensures campaigns can drive results when they are most impactful and most fruitful for end advertisers.