Is Your Agency Ready for Big Data?
By Mike Carlton
The merger of Publicis and Omnicom is the talk of the industry. Advertising Age called it “Adland’s Biggest Bang.” It consolidates the second and third largest agency holding companies. Both of which have already been bulking up their big data capabilities with new and expanded data intensive units. Big data was announced as one of the key reasons for this merger. As Maurice Levy, the head of Publicis is reported to have said, (we will) “build something quite incredible in terms of crunching the data.”
What is Big Data?
So, what are we talking about here? The whole notion of big data is really quite recent. And like any emerging concept, accepted definitions and successful marketing applications of big data are still a bit elusive.
Wikipedia defines big data as; “A collection of data sets so large and complex that it becomes difficult to process using on-hand database management tools.”
In a more down-to-earth way Oracle has this to say about it; “The term big data draws a lot of attention, but behind the hype there's a simple story. For decades, companies have been making business decisions based on transactional data stored in relational databases. Beyond that critical data, however, is a potential treasure trove of less structured data: weblogs, social media, email, sensors, and photographs that can be mined for useful information.”
Bottom line is that big data can provide marketers with more information about customers and prospects than they have ever imagined having before. And not just groups of people but right down to each individual.
The potential of big data has prompted industry gurus to predict that big data has put advertising agencies on a collision course with data firms and that the future of agencies as we know them is in serious question.
That may be a bit of hyperbole, but the implications for the agency business are huge. Clearly Publicis and Omnicom are betting the farm on the importance of big data in their future.
Well known names like Google and Facebook have accumulated tons of data about individuals and are not hesitant to use it. Neither are other technology giants. And new players like Acxiom and Epsilon are helping marketers to better manage their customer knowledge data. Collectively they – and hundreds of other entrepreneurial start-ups - are diverting to themselves significant amounts of marketer spending that used to go through agencies for advertising.
A Deep Breath
Against all this background noise – and the prophecies of doom - perhaps it’s time to step back and pause to think more deeply about what the implications of all this really are.
And more importantly, how advertising agencies can turn big data from a threat to an opportunity.
A Key Principle
Let’s start here. From the beginning of time, one of the most important principles of commerce has been;
The better you know the customer, the more effective you can be.
Now there is nothing startling in that principle. It’s simple and logical. We all know it. And we all use it every day. But it is so intuitive and so ubiquitous that it’s easy to forget. At its core, big data is nothing more than knowing the customer better than we have ever known her before.
That is the important message of big data. Improved understanding of the behaviors of each and every customer can be known by the marketer in more detail and with more precision than many of us ever thought possible.
Precision, Relevancy and Efficiency
Big data is clearly a game changer. By knowing each customer better we can dramatically improve the precision, relevancy and efficiency of every commercial interaction. So the hype is not without great substance.
Yet at the same time, big data is not an end all. It should not be given credit for more than it can deliver.
There have always been two basic steps in effective market communications or any other form of commercial interaction.
The first step is determining what to say.
And second step is creating how to say it.
Whether they have codified this practice or not, all good agencies have practiced this two-step approach. Historically it was the job of the account service and planning working with the client to determine with as much precision as possible what to say. Often they did not have much hard data to support their point of view, but any effective outcome required as tight a focus as possible in establishing what to say.
Then came the creative folks. Their job was to create the best way to convey that message. The how to say it. Knowing that the recipient of any message would react most favorably if the message was more than just factual but actually touched him emotionally. So that the outcome was inspirational.
Whether overtly or not, most great advertising evolved from these two steps. There was as much precision as possible in determining what to say and the best of creativity in determining how to say it. Without both steps working in harmony the message could be off target or just plain boring. In either case ineffective and a waste of the marketer’s money.
The Role of Big Data
Obviously, big data goes right to the heart of the precision and relevancy of what to say. Not only when addressing large groups of customers, but right down to what to say to each individual human being. This benefit is huge.
And because it so dramatically improves targeting, big data can improve the economic efficiency of virtually all forms of commercial interaction. The financial benefit it can deliver can far outweigh the costs associated with employing it.
While big data is transformational, we should not attribute to it benefits it cannot deliver. Big data is not very good at addressing the second step, creating how to say it in an inspirational way.
Data, by definition, are historic. Even facts reported in real time are historic. And no fact, nor any piece of information, can accurately predict the future. Yet the marketer’s job – aided by its agency - is to shape future consumer behaviors.
Think about that for a minute. All this data is really about what was. And the success of marketing initiatives will be measured by future outcomes.
That raises the question about big data’s role in predictive modeling.
Big data can help a lot. But its ability to predict future behaviors is still primitive. Just ask yourself how much faith you have in next week’s weather forecast. A forecast based on the best data, computers, systems and software that money can buy. Then apply that same level of reliability to most consumer behavior predictions.
The Problem with People
Like the constant, yet often subtle, changing winds and temperatures that make weather forecasting so difficult, people are constantly changing, too. In ways we never fully understand. Nor can accurately measure.
If you doubt this, ask yourself why data-based election predictions are constantly changing and why accurately predicting political outcomes is so difficult. Those pesky people out there just keep changing their opinions and their behaviors. They just won’t sit still long enough for reliable predictive models to be constructed. Kind of like trying to build a house on ever-shifting sand.
How people behaved in the past and how they will behave in the future is not easily reconciled by quantitative means.
No matter how sophisticated the computer or system or big data set.
Here’s where the real strength of agencies comes in. A good creative agency can take the precise what to say brief and communicate it in a way that touches the individual at an emotional level and leaves them inspired.
Data driven marketing automation can’t do that. It can create a workman like on-target message. A message that is precisely focused, perfectly relevant and highly efficient. But it is a message not likely to touch the recipient’s emotions in any kind of sophisticated humanistic way. Nor is it likely to inspire. While marketing automation is a powerful tool, computer generated content set on auto-pilot can only go so far.
Only an intuitive creative human mind can raise what to say input to an inspirational level.
And as we all know, an inspirational message can have an economic multiple many times greater than an on-target but mundane message.
Elevating the how to say it to an inspirational level through creative leverage is the magic dust of a good advertising agency. Always has been. And always will be.
Psychology and Statistics
Over the years when young folks hoping to enter the agency business asked me what academic areas they should include in their studies I always said the two most important are psychology and statistics. While that may sound like a strange combination, it goes right to the heart of what advertising is all about.
The basic purpose of advertising is to change the behaviors of consumers. And a thorough grasp of psychology can help us understand why people behave as they do. Obviously critical knowledge in creating inspirational messaging.
While quite a different field of knowledge, statistics can help us understand the likelihood of people adopting the behaviors the marketer desires. And then quantifying the potential return on investment of the advertising spend.
Most good agencies are populated by people strong in psychology, but not many have strength in statistics.
Thus now is clearly the time to strengthen the statistical side of the business. Not to become geeks in big data but to become masters at taking big data information about customers and using it appropriately in crafting the creative content that will touch the emotions of the customer.
Big Data and the Advertising Agency
So, what does this all mean for an advertising agency? And how can big data become a great friend?
Here are some things worth thinking about:
Senior managers at many marketers are very much into big data. So any agency that resists or pooh-poohs the benefits of big data puts itself on a course for out the door. While engaging clients by exploring their big data opportunities can help move an agency a rung or two back up the food chain.
Staff apprehension can also be an obstacle to embracing big data. In some respects it is an alien concept intruding on a fairly comfortable employment model. Smart agency leadership will need to make sure that their own people understand that big data can make their work more effective and thus more valuable.
Big data will bring in an increasing number of new client side people. Folks that are trained in technology but are not necessarily strong in marketing. Big data is technology based. And like all disciplines it has its own terminology and jargon. Agency people need to learn that terminology and jargon so they can communicate effectively with their opposite numbers.
This will likely mean reaching out to engage with new partners who can develop and refine the what to say portion of the equation while your agency is focusing on the how to say it challenge.
An agency that can build a bridge between big data’s ability to determine what to say and the agency’s ability to create the most effective way to say it is of great value. It is worth big money.
So make sure that you are fairly compensated for this important contribution. This is high value work that should command much more than a run-of-the-mill hourly rate. And if you are not already into value based compensation this is a good place to start.
Many years ago John Wanamaker, one of advertising’s greats, said, "I know that half of my advertising doesn't work. The problem is, I don't know which half."
Big data can improve those odds.