Programmatic in the ‘Second Machine Age’


frostprioleaurevisedData-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 Frost Prioleau, CEO and co-founder of Simpli.fi.

Programmatic advertising continues to grow faster than analysts’ expectations, leaving industry pundits to continually raise growth predictions made just months before.

While this acceleration has been impressive, the industry is just getting started with even bigger gains in computer power to come – if we use the history of chess as a guide.

In the book, “The Second Machine Age: Work, Progress, and Prosperity in the Time of Brilliant Technologies,” authors Erik Brynjolfsson and Andrew McAfee note how a king, impressed by the game of chess, invited its inventor to calculate his own reward for his creativity.

The inventor asked the king for rice that was doled out in the following manner: He put one grain of rice on the first square of the board, two grains on the next square, four on the next, eight on the next square and so on.

On the second half of the chessboard, in squares 33 through 64, the nature of extended exponential growth really kicked in. The amount of rice needed for the 64th cycle was 210 billion tons, which would require a growing area twice the Earth’s surface area.

Ad tech, like many fields, benefits from the exponential growth in computing power as defined by Moore’s Law. While the growth in computing power to date has been extraordinary, we’ve still seen fewer than 32 cycles. This means we are only halfway through the proverbial chessboard, and even if the pace of doubling slows to every two or 2 1/2 years, the big gains in computing power are yet to come.

I see several beneficial ways in which a “Second Machine Age” of automation could affect the programmatic ad space.

Fully Automated Media Planning

Just as automation is bringing more accurate diagnoses in the world of medicine, machines are already getting better at media planning than humans. As good as humans are at pattern recognition, we cannot absorb and compute anywhere near the amount of data required to understand the interactions between media placements, user characteristics, intent signals and time of day, among other inputs.

Ad tech is already generating media plans by analyzing the behaviors of advertisers’ existing customers. Media plans will become much more detailed and target individual prospects, instead of pre-packaged media plans for thousands or millions of prospects at a time. The individual plans will optimize on the fly as new data is collected, delivering better ROI for the advertiser.

One-To-One Marketing Moves Up The Funnel

The future of programmatic advertising is personal, connected and meaningful to customers. Creating an effective, continually optimized audience from automated systems is only one part of the opportunity. The other part is delivering customized creatives to users on a one-to-one basis.

Currently, lower-funnel site retargeting campaigns make extensive use of dynamic creatives customized to display products previously viewed by a user. However, in upper-funnel prospecting campaigns, broad segments of prospects are typically served identical creatives.

Coupled with automated media planning, customized ads can be targeted to individual users who have never visited the advertiser’s website based on their search history, site visitations, geolocation history, time of day and other anonymized data.

Automated Analytics And Insights

We all know how time-intensive analytics and reporting can be. Extracting meaningful insights out of the massive volumes of data available in a programmatic campaign can be a massive resource pit.

Automated data visualization is already giving back countless hours to personnel who wrestle Excel files, input data and format charts. In the future, machines will continue this trend of automatically determining the most relevant data to highlight and presenting it in the format that is best for easily digesting the appropriate insights.

Due to increased automation, these insights are increasingly becoming available on demand at any point throughout the campaign instead of having to wait until the end of the week, month or campaign. Additionally, on-demand customized reports that query petabyte-level data clusters to drill down on specific issues are becoming more common, enabling advertisers to pull their own custom reports at will instead of having to request them from a platform or vendor.

Fully Automated Ad Tech

The future of programmatic will likely follow a progression similar to automation in cars. In the early days of car ownership, drivers needed to go through multiple processes to start the car, with the understanding that these processes may vary under different conditions. Today, drivers expect a car to start immediately and run well in all conditions. This is because the intelligence has been built into cars that enable them to self-diagnose and self-correct.

That same intelligence is being built into programmatic platforms. In the future, systems will automatically determine the best ways to set up and manage campaigns to meet the advertiser’s goals.

Today, there are many human processes aided by machines. In the future, it will be the reverse.

Follow Frost Prioleau (@phrossed), Simpli.fi (@Simpli_fi) 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>