No More Optimization For Optimization’s Sake

Data-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.

Today’s column is written by April Brown, vice president of marketing analytics architecture at Origami Logic.

Programmatic technology has democratized access to data-driven advertising. The average marketer has access to unprecedented levels of raw information that probes consumers’ needs, wants and desires. Obtaining and operationalizing data is no longer the challenge. The question before us now is how to best optimize our advertising against what the data is telling us, in a way that balances both efficiency and effectiveness.

Today’s marketers typically emphasize one to the detriment of the other. They will continue to do so as long as media optimization occurs in a vacuum – optimization for optimization’s sake. The true context of a campaign – its measurable impact on the target audience and the discernible effect on business outcomes – is often cast aside, leaving marketers to optimize campaigns according to one-dimensional internal logic.

That leaves a yawning gap between what data can do to maximize efficiency and what data can do to improve business results. It’s time we look critically at the difference and work to close the gap.

Efficiency without effectiveness

Programmatic advertising was fundamentally developed to maximize efficiency of impression-level spend in near real time. Programmatic platforms are amazing trading mechanisms and – bonus for the inquisitive marketer! – give off huge exhaust plumes of luscious data. However, the kind of supply-chain optimization that programmatic does so well lacks the contextual information necessary for marketers to optimize with clarity and understanding.

Supply-chain optimization is necessary – but not sufficient. Efficiency without effectiveness is necessary but not sufficient.

As the marketer’s marketer Peter Drucker said, “Efficiency is doing things right; effectiveness is doing the right things.”

Who cares if you get a great CPM, but your conversions are so low that your marketing budget gets cut? Optimizing to the right balance of efficiency and effectiveness requires a nuanced understanding of business objectives and the contextual factors defining business success.

Many times the stated objectives for a given campaign, for example, are as simple as, “I ran an awareness campaign, I have reached targets in terms of volume and I need to get it for a good price.” Internally consistent, yes. But responsible? Not exactly.

We must go further. We need to look at not only how many people were reached and the cost of reaching them, but who converted – conversion in the broadest sense – and what creative piqued their interest. Did conversion vary by timing, age, geography?

Optimizing against the CPM alone can generate efficiency. But look deeper at a very modest amount of data that provided contextual value to CPM, and you can see that some audiences will perform much better than others, some publishers better than others and that some creative formats outshine the rest.

With a little slicing and dicing, this can be achieved with a minimum of metrics: CPMr, CTR, CVR and frequency. Example slices of the data might be audience groups, geography, creative format and message and publisher.

This added layer of optimization is achievable, but only if marketers stop assuming the machine will run itself. We need to think creatively and critically, and be intentional in how we map our business context into the campaign.

Context is the watchword

One way to keep the vision big but the execution manageable is to start by scaling the analytics strategy down to the minimum viable handful of metrics that have the potential maximum business impact.

Slice and dice your chosen handful of metrics to help you optimize against audience, creative, inventory and investment areas. Think through the levers you might pull while your campaign is in flight. That may include, for example, swapping creative, throttling back spend, altering your bidding strategy or increasing spend for high-performing audiences.

If you take this approach, it is likely that in the rush of warm fuzzies you get from delivering business value quickly, the tendency will be to rush toward more – more data, more analysis, more granularity and more frequency. Marketing data presents thousands of potential KPIs and myriad potential ways to slice and dice the data. It doesn’t take long, though, before this thrill is replaced by a persistent suspicion that whatever one is measuring is not enough –not granular enough, not often enough, overly simplified or not even the right view for the “dart thrower” back at the office.

All good, though. As long as you are disciplined and your exploration delivers context, you can – and should – enjoy the ride.

Make context your watchword and be of good cheer. Your efforts allow you to move one step further on the journey to greater business impact for your company and a better customer experience for your brand.

Follow Origami Logic (@OrigamiLogic) and AdExchanger (@adexchanger) on Twitter.

 

Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>