Ian Pearman says 2013 could be the year of the algorithm, when data scientists become as vital to agencies as creatives.
"Millpond Britain", an economist friend calls it – the stagnant stillness created by low expectations. Businesses and households alike stuck on pause, too nervous to spend or invest. So nothing happens. Nothing changes.
And while they venture as far as a low single-digit forecast here and there, the media commentators are equally pessimistic. Spends may rise by 2 or 3 per cent at best, but no-one is betting the farm on much beyond that.
So the logic should direct us towards pessimism: 12 long months of grinding, torturous scrimping and saving. Agencies will get even more vicious in fighting for share, brokers will leverage this to their advantage and procurement-led pitches will be at the mercy of the dumbest competitor. As always, the strongest agencies will hold their ground, the weakest will slide. As Warren Buffet said: "It’s only when the tide is out that you can see who’s been swimming without any shorts on."
And that same logic directs agency managers to plan for the worst and hope for the best. We write plans that are broadly flat, with as much flexibility built in as possible – as equally ready to cut or thrust depending on the economic smoke signals.
But the assumption that a flat-lining economy and a flatlining media market means a flatlining ad industry is suspect.
Long broken has been the link between media spend and agency income. There was some causal correlation 20 years ago in the heady days of commission, but it is hard to see the link as anything but casual now. If the move to payment by hours created the cracks in the connection, the growing importance of earned media smashed it. As Peter Field’s fabulous work with the IPA showed, social media means the commercial impact of highly creative campaigns is greater than ever. The hypothesis is simple: campaigns that are worthy of creative awards are much more likely to be shared and are even more likely to be commercially successful. When so much of the impact of our industry’s ideas is becoming a function of earned rather than bought media, the real metric for our own health should be how successfully we manage and monetise our contribution to that effect.
2013 could and should be the year when ad agencies get a real grip on charging their clients for earned media: for identifying the right targets, for the management of those audiences, for the creation of specific content for those audiences and for the measurement of its impact. The media impact might be "free", but our contribution to it shouldn’t be.
That means bringing the competency in-house. Only then we will be able to do it properly and retain the benefits. Social agencies have made a very fine business monetising the sticky ideas created inside ad agencies while dodging the heavy costs of creative origination. Ad agencies should and will lose patience with these free riders and that will drive a new wave of consolidation between competencies, supercharged by clients’ pecuniary interest in reducing agency rosters.
And the earned media competency is becoming increasingly technical. In 2013, more than ever, it won’t be media that sits between brands and consumers but algorithms. Brands and their ad agencies need to become much better at planning for the algorithms that control much of our digitally connected media. To get found on Google, a brand has to fully understand and build around the monster that is the Penguin formula; to get shared on Facebook, work has to beat EdgeRank, which recently limited reach to 16 per cent of a fan base. If a bunch of algorithms can cause the Dow to drop 10 per cent in 20 minutes and it takes months to work out how and why (and there are 2,000 physicists on Wall Street), the risks to little old brands of not playing the equations properly are terrifying.
So we have to invite the data scientists in. The people who can understand, write to and negotiate with these mathematical formulas that decide which bits of data are relevant are going to become as valuable and as essential as the creative teams who come up with the ideas.
And perhaps this will be the year that the formulas start affecting the ideas themselves. Epagogix is a British company that runs algorithms on Hollywood film scripts to determine if they will be successful based on a multitude of variables from location to cast. The company claims that, in one case, it saved a major studio hundreds of millions of pounds on a film that was projected to make £180 million, but, after its analysis, it turned out to be only £30 million. Film-writers of the future beware. It’s not the audience you have to write for, it’s the script-testing algorithm. Will 2013 be the year that an advertising research company suggests a similar service?
That millpond’s not looking so still.
Ian Pearman is the chief executive at Abbott Mead Vickers BBDO
This article was first published on