Like millions of other viewers, this week I find myself mourning the “End of an Era” after the seventh and final season of AMC’s Mad Men wrapped last Sunday night. Throughout the series (the show’s timeline is 1960-1970) we watched the characters grow and change—but even more dramatic shifts happened in the workplace itself. On the social side, feminism advances women from secretaries to partners and copy chiefs, and the civil rights movement integrates the agency’s employees. New technology makes waves: a Xerox machine and eventually, Sterling Cooper’s first computer, IBM System/360. The relevant quote that caught my attention, however, was in a recent episode where (minor spoiler alert) everyone is relocating to McCann Erickson, and Harry Crane says: “McCann is Mission Control. Statisticians, programmers, five men and ten women just handling data!”

This made me wonder what Peggy, Don, and the crew could do with Big Data and analytics tools at their disposal. I’ve heard the claim that “every ad agency is just one phone call away from being fired.” We certainly see this on Mad Men—accounts come and go like the wind, sometimes for the most insignificant of reasons. Oftentimes the clients are inexplicably unsatisfied with the campaign, don’t see immediate benefits, or capriciously decide their work is better suited for another agency (hardly decisions based on data). In the past, it has been incredibly difficult to measure the impact of advertising and marketing initiatives, but this is where predictive analytics is bringing change as big as Roger Sterling’s early 70s mustache.

Today, advertising agencies, marketers, and companies can have real-time analysis on the impact of their campaigns, and even predict how well a new campaign will fair. Marketshare, a marketing analytics firm, discovered that one client’s YouTube ads, 6% of their budget, were twice as effective in generating online searches that led to purchases than their TV ads, which were eating up 85% of the budget.[1] The client rearranged their ad spend dollars—without increasing the budget by a single dime— and the result was a 9% sales increase. McKinsey found that data-driven marketing decisions, like this example, can increase companies’ marketing ROI by 15-20%.[2] In addition to bringing in new customers, analytics can tell you exactly who these customers are. The technology allows organizations to develop profiles of their various customers and determine with a high level of accuracy how likely any particular customer is to make a purchase, and if so, what product they are most likely to purchase. This allows for much more targeted and personal advertising, especially indirect mail and online advertising. Companies can learn where to spend the most resources and where to spend the least—why spring for the expensive, spiral-bound brochure if that customer has little to no interest in the product in the first place? Send the fancy materials to the people with a high probability of purchasing, instead.

Analytics may be even more impactful in the social media marketing sphere. 500 million tweets are sent on Twitter each day[3], and 2.5 billion pieces of digital content are shared on Facebook. However, while there are quantified “likes” and “shares” on these sites, more than 90% of social media data is unstructured[4]—meaning the true value lies in text, viral videos and “hashtags,” which are rather difficult to analyze in a spreadsheet. One can assume this means that without the right tools to process and analyze unstructured data, marketers are missing out on the majority of important customer insights that come from social media. Sophisticated Big Data tools like Hadoop are now making it possible to analyze unstructured data and visualize what people are saying—right from the moment an ad airs. This information can be used to make real-time changes (e.g. pull an unfavorable ad before irreversible damage is done or increase social media promotions when sentiment is favorable). It can also be used a variable in predictive analytics modeling to determine the effect social media sentiment had, or will have, on sales or other metrics on any given day.

Maybe the best creative ideas for ads will still come from human minds and emotions—it’s hard to imagine a computer replacing Don Draper’s Zen-revelations on a California beach. But when it comes to making better media and spending decisions, understanding customers on a deeper level, and impressing clients—Big Data and analytics are the new advertising tools in town. Who knows, maybe there will be a futuristic Mad Men spinoff where Pete Campbell is going after accounts with data visualizations up his sleeve. (Unlikely— but I’ll still keep wishing for another season).


[1] https://hbr.org/2013/03/advertising-analytics-20

[2] http://www.forbes.com/sites/mckinsey/2013/07/22/big-data-analytics-and-the-future-of-marketing-sales/

[3] http://www.internetlivestats.com/twitter-statistics/

[4] http://www.businessinsider.com/social-medias-big-data-future-2014-2