In the first two posts on Advertising Process Control (here and here), I explained that there was great value in applying process control mechanisms to your advertising just as you do to things like average days outstanding for accounts receivable and customer satisfaction, and laid out what achieving Advertising Process Control requires in general terms.
Here, I discuss the actual steps that should be applied to the many non-creative areas that advertisers, publishers and agencies could and should work to control better to consistently improve their performance across advertising campaigns. (What makes for good creative is outside the scope of this series of blogs.)
Not surprisingly, the elements that make for a good advertising process across multiple campaigns are those that take the elements that make for a good single campaign and replicate them. The key to the strength of your advertising process control is whether you begin to see the results across campaigns that you can produce for individual campaigns. Are you increasing brand lift? Are you reaching the right people? Did you drive sales?
There is no single, absolute formula for advertising effectiveness – which is only to say that the science of advertising effectiveness is not complete. However, there are common elements that, when measured and optimized consistently, do result in consistently higher performance.
Suppose your ads consistently—or inconsistently—perform poorly. How should you proceed? In a 2013 Nielsen survey, nearly all brand marketers indicated that "having everyone collaborating directly to produce the best possible outcome for the campaign" is important to maximizing ROI for online brand advertising initiatives. There is every reason to suggest exactly the same is true for offline advertising initiatives.
But what should a beginner in the science of advertising process control have everyone collaborate directly over? The effectiveness of any given campaign is a function of whether or not you're reaching the right audience through the right medium with the right creative at the ideal frequency at the ideal places offered by that medium—e.g., shows for offline, websites for online, and apps for mobile.
Obviously, any weakness in any part of the process flows through to weaken the overall result.
Advertising process control is a new idea, and busy company departments don't make a lot of time for adjusting their core processes. Despite the fact that large advertisers might spend billions on advertising annually, it's almost universally the case that companies are not yet analyzing all the elements of an advertising campaign reliably. If they are not conducting the analysis fully, it's not surprising that they are not yet conducting the meta-analysis—the analysis, that is, of the analysis.
Who ought to be driving advertising process control in your company? More than likely, it’s someone from either senior marketing management or someone with access to all the data and the savvy to analyze the variance properly. It is senior management's job to continually improve the basic processes within the company given that most of the folks running the processes don't have the time or perspective to see the big picture and drive the appropriate change. This may be particularly true about advertising process variance measures, which can highlight various uncomfortable comparisons, impact strategy or even suggest organizational change.
Recently, a particular advertiser examined the performance of all campaigns for a particular brand. Leveraging the variability of that performance, he was able to identify efficiencies in the process: he discovered two particular campaign groupings. One grouping consistently had a high brand lift; the other set of campaigns had a consistently average performance. The first group had in common a great media planner who would frequently optimize the campaigns. The insight for the lower performing group was that most of these campaigns had consistently used an underperforming creative.
The key element in this framework is to analyze the advertising process and identify the variables that consistently produce good results rather than looking at the campaigns individually.
Today, we have reams of granular data from helping marketers make smarter decisions to deliver more impactful campaigns. That's the heart of optimization. But to understand how to reproduce good results on a consistent basis, advertisers need to think big picture, which is what ad process control is all about: aggregating the results from individual campaigns to derive broader insights into those variables that affect performance (ad placement, frequency, on-target reach, etc.). Once you know that, you can manage for those factors and drive strategy to consistently deliver the desired outcome.
While ad process control can be used in many ways, our recent analysis of the performance of one client's digital ad campaigns across a variety of publisher types serves as an example. The highest campaign performance was found on Special Interest and e-commerce sites, and this performance was found to be relatively consistent for the entire set of campaigns. For this advertiser, one overarching takeaway is that precision marketing at the contextual level is a good practice.
The best practices for marketers are pretty simple. First, measure your campaigns using metrics that matter to you. Whether that's brand lift, sales lift or something else. Second, don't forget to look at the big picture. It's so easy to get hung up on one campaign and then quickly move on to the next. Optimizing a specific campaign might require all your time and effort, and for good reason, but taking a step back might lead to bigger returns in the long run.
Finally, measure the variance with which you are producing your results at every step of the process. That way you can improve the process of creating and delivering effective and efficient advertising.