"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 Tom Triscari, co-founder and managing partner at Labmatik.
It’s that time of year for annual reviews, praises, raises, reprimands and firings.
Since marketers don’t actually buy ad tech – they hire it to get a job done – why not put programmatic through the same rigor?
So, let’s give it a persona and a name – Programmatic Joe – and put him through an annual review.
Some employees like Joe can’t wait for their annual review, while others dread that their underachievement will be exposed. Others are simply indifferent and happy to go through the motions on their way to another year with a steady paycheck and health benefits. Joe is no different — the status quo arrangement has worked well for him.
The 9-box method is the most appropriate to assess Programmatic Joe. The 9-box system rates employees based on past performance and also on potential, which is a probability of future performance. A manager scores an employee as low, medium or high along each dimension.
Good employees land in the upper-right box because they have demonstrated great past performance – team player, knowledge, skill, intellectual horsepower – and show great promise for future performance. They should expect a nice raise, perhaps some stock options and maybe a promotion.
“Good enough” employees end up somewhere in the middle. They are necessary cogs in the machinery who get routine stuff done and occasionally show signs of promise. Their main obstacle is learning how to discover and enhance their strengths instead of trying in vain to overcome weaknesses. Getting past their own hurdles will make their 9-box scores shine next year.
Then there is the lowly, bottom-left quadrant: subpar performance, teammates think they are unreliable and they show no potential to improve. Is it you? Are you just plain old lazy, not smart enough or perhaps the company or industry is not the right fit? Either way, something has to change, and it is most likely going to be you. Is this Programmatic Joe?
Rating Programmatic Joe with the 9-box model
Perhaps the first thing the boss would do is look at Programmatic Joe’s annual review history. For the 10 years Joe has been on the job, he has received hefty annual raises every year. In fact, his annual review looks the same year after year:
- High performance (based on measures that are not actually real)
- High potential (even though reliable fundamentals have never been established)
- 25% average annual raise (in the form of increased budget allocation)
Programmatic Joe’s job objective would read something like, “Be a positive contributor as one important component in our media-mix model.”
His file provides more detail:
“We allocate 20% of total paid media to programmatic, 10% to search, 20% to social and the remaining 50% split across traditional TV, out of home (OOH), radio, etc. Our media-mix multiple regression model will measure Programmatic Joe’s marginal contribution to sales or brand outcomes and also estimate his effect on the other channels.”
However, during his review, Programmatic Joe takes a very different point of view. His job objective comes up every year and the same outcome always prevails. Programmatic Joe agrees with his boss that linking performance to real business outcomes makes total sense, but for all practical purposes, his real day-to-day job involves spending significant media budget at the lowest possible CPM. Joe’s real job is to drag down the overall CPM average in order to reach synthetic targets set by procurement and mandated to agencies, which are only too eager to put Joe to work.
Luckily for Joe, his boss is reasonable and recognizes that he is right. Instead of getting a low score with his media-mix model contribution used as a measuring stick, Joe is given high performance scores and let loose to reach his enormous spending potential, even if the ads are never seen by anyone and can’t possibly move the needle on sales.
Getting more serious about Joe in 2019
Programmatic Joe’s annual review should go differently this year. With $70 billion in annual programmatic spend and tentacles quickly reaching into connected TV, audio and OOH, the boss is taking a conscious step back before considering such a big promotion.
The boss realizes that she has been Joe’s biggest fan for the last 10 years, so now she is stuck with him. As much as she believes in the promise to get it right, she is also worried about his lack of fundamentals. So her best move is to put programmatic on an performance improvement plan – it’s up or out for Joe.
The boss hears that programmatic viewability may not be as high as she was led to believe, so that’s the first thing Joe must fix. The boss also wants to only get billed for downloaded ads that actually render on a page, instead of paying for served ads that don’t.
The boss wants Joe to achieve other stretch goals, such as contracting directly with supply-side platforms, writing all vendor data to a blockchain layer or other normalized data lake and reducing identity error rates by rethinking the shortcomings of their data management platform. Who knows, maybe AI is the answer for Joe, if only the sales pitch could turn the dream into a viable use case with clear benefits.
Joe will no longer be employed to buy cheap or valueless CPMs. The boss has also hired a dedicated statistician whose only job is to work with the media mix team and measure programmatic’s effect on real business outcomes.
If Joe succeeds it will be because he finally established reliable fundamentals and proved he can move the needle. Only then should Joe get that annual raise and big promo.