Books reviewed by Raphael Badagliacca
Frenemies: The Epic Disruption of the Ad Business (and Everything Else)
by Ken Auletta (Penguin Press, 2018)
This book, written by the longtime media and communications writer for The New Yorker, should on the surface be of great interest to any leader of an association dependent to any degree on advertising, which, along with trade show income, is a major source of non-dues revenue. But I’ll go further to say that it should be of interest to any leader of any organization of any type because in the age of the internet every organization is a media organization and in the words of the author advertising is media’s “fuel supply.”
This is not an anti-advertising book. Right at the outset, the author states how advertising and marketing are necessities in a free society.
Then Ken Auletta takes us on what could only be called an encyclopedic history of the transitions the advertising industry has gone through to bring us to the present -- where the word he has chosen for his title describes the relationship between advertising agencies and the internet giants whose unprecedented fortunes come from advertising income – Facebook and Google.
Like so many things these days, it all comes down to data and who owns it.
There is a not-so-gentle reminder here, and another compelling reason to open this book, that associations and other non-profits are the guardians of their constituents’ data and that their continued existence is dependent on maintaining the trust of those constituents.
In nearly 350 pages peppered with the names of notable individuals and companies and where they figure in the current advertising matrix, with numerous nostalgic comparisons to the days of Mad Men, carefully articulated power distinctions like that between creative and media agencies, and what import the ownership of name agencies by publicly traded companies has had on the business – a loss of trust – it still all comes down to one word – data.
The internet transformed advertising forever by turning it into an outcome numbers game, although you will also find in this book the arguments of critics that as much as 60% of the clicking going on is being done not by humans, but by bots. True or not, internet advertising is now perceived as being much less about guesswork than it was in the days of Don Draper. It’s not the creative presentation that wins the account but the data plan.
“Walled gardens” are the two words used to describe the data Facebook and Google will not share with the agencies who bring them their ads. This is the data that allows for the targeting of individuals. It’s the data that would allow agencies to go to other media platforms knowing what is working and what is not, something which neither Facebook nor Google wants to see happen.
The internet giants each strive to become the entire internet, to conduct business with as few intermediaries as need be mustered, the organizing principle of disruption, an ominous trend for advertising agencies and one that promises to further transform advertising as we know it.
Hello World: Being Human in the Age of Algorithms
Hannah Fry (W.W. Norton & Company, 2018)
Here is the simplest definition of an algorithm: “A step-by-step way to solve a problem.”
One of the advantages of reading a number of non-fiction books about a single subject, even if that subject is as broad as what it means to experience the world today, is how they talk to each other.
So, in Homo Deus by Yuval Noah Harari, which I reviewed earlier in this column, the author explained to us what an algorithm was by giving the example of a vending machine. The machine is programmed to follow a series of steps that include receiving your coins knowing what button you pressed and delivering the can of soda you desire. He went on to describe a future in which algorithms would also be designed to walk up to the machine and press the button. He also made the off-handed comment that most advertising these days has a single client – the latest search engine algorithm designed by Google, something I expect the author of the other book reviewed this month would not argue with too much. Despite all the information on websites and in online forums, we will always need books written by intelligent authors to slow things and put them together for us.
That is exactly what Hannah Fry has done in Hello World. She explains to us that we are surrounded by algorithms and illustrates for us how much of our decision-making we have consigned to these mathematical constructions. In so doing, she also cautions us not to trust them as much as we do.
She distinguishes between rule-based algorithms and machine-based algorithms.
She gives vivid illustrations of examples where human observation correctly over-ruled what the algorithm had decided, the most dramatic of which was the decision made by a Soviet military officer not to report what the algorithm mistakenly told him – that five nuclear missiles were on the way to his country from the United States. There are many catastrophic stories where people blindly followed the directions of the algorithm controlling their GPS systems to drive off cliffs or into rivers.
She reports our tendency to view the accuracy of algorithms in black and white, implicating her own behavior. Said differently, we tend either to not trust at all or trust completely what the algorithm decides, usually in that order. At first, we doubt the GPS is taking us the right way because it feels unfamiliar, so we discount its accuracy, but once it gets us there, we are sold, and give up our critical faculties for all rides going forward.
Her chapters cover the great degree to which algorithms are already making decisions for us in the areas of data, justice, medicine, cars, crimes and art. Her in-depth discussion of data, privacy, and how much algorithms are used to track us, decode us, and know our preferences is at once impressive and alarming. Every association leader, take note.
The future she would like to see is best expressed in her chapter on medicine where she discusses how biopsied cells are read for signs of breast cancer. If pathologists had only five slides to review a day, they would be 100% accurate in their analyses. But that’s just not the reality. The best arrangement would be for the powerful pattern recognition capabilities that an algorithm could provide to do the grunt work, and then have a human make the final decision on the slides that have been singled out – a collaboration between algorithm and human.