Big Data and analytics unlock the truth about marketing’s impact on growth.
Measuring the impact of marketing on revenue (or profit) growth has always been a delicate balance of science and common sense. We need the science to keep common sense from devolving into conventional wisdom. We need the common sense to guide our application of science in commercially pragmatic and material ways.
In times (like these) where the marketing ecosystem is evolving so rapidly and marketing tactics are splintering into hundreds of customer/prospect touchpoints, this delicate balance is continually threatened by several strong forces.
First, the tendency to hang on too long to scientific methods that have worked well in the past slows our measurement progress and frustrates executives seeking to find and exploit growth opportunities. This also undermines the credibility of science in general amongst those who exist “on the margins” between wanting to believe the science but torn to trust their experience and intuition – a sub-segment which can represent 50% or more of any senior management team.
Second, the passionate advocates of the newer tactics thrive on the high marginal returns of the next dollar spent off a low base, tirelessly arguing for more funds justified by narrowly framed direct attribution analyses. This sets up conflict within the organization between the managers of the “new” and the overseers of the “old” (who often control the largest percentages of the budget). Each side often hires specialized agencies and consultants to help them maximize the impact of their tactical domain, thereby setting off a competition that only serves to obfuscate the true drivers of growth beneath layers of rationalization and justification.
Third, these tactical fractures are often accompanied or caused by technological shifts that further complicate the resource allocation process. In the most recent case, the advent of “big data”, cloud computing, and social networks have armed all sides with both the analytical and emotion-plucking weapons necessary to overwhelm executives with compelling cases for charging even faster into an environment characterized mainly by higher levels of uncertainty.
All of this has been shown to cause three management decision-response patterns, all of which ultimately hurt the company’s growth prospects:
- The Aggressor – plows much more money into the newer tactics in pursuit of fast growth, failing to realize that there are few barriers-to-entry behind them. This pattern, even when initially successful, tends to produce poorer results when the longer-term portion of the payback stream is disrupted by the blunting efforts of the bumbling masses who follow, cluttering the environment and in some cases poisoning the regulatory well along the way.
- The Resistor – spends more time arguing that the “fad” will pass than they do legitimately exploring how it might be embraced and used to drive growth, often resulting in a decline in customer acquisition and a commensurate increase in retention costs as new channels reach both prospects and customers effectively.
- The Ostrich – analyzes the situation to the Nth degree, hoping for a clear answer to emerge to guide their actions, missing the opportunities to test and experiment to gain further expertise and informed judgment.
Few companies demonstrate the measured, balanced approach to embracing the new ecosystem thru controlled exposure and continual risk/reward assessment.
This is where science has failed the marketing discipline in the past few years. Science has been too slow to adapt to the fundamental shifts in the marketing ecosystem, and has left a void that is forcing companies to “triangulate” on marketing effectiveness using multiple limited tools like Google analytics, digital attribution engines, brand trackers, and traditional mix models. It’s a totally reasonable approach by managers trying desperately to make sense of imperfect information. But it is totally un-reliable, owing to the substantial margin-of-error in the size of the triangle, not to mention the availability bias of the individual methodological points defining the triangle.
Worse yet, it is totally un-necessary. The very technological advances which complicate the problem offer the key to its solution.
Big Data can be gathered more cost effectively than ever before and in more timely ways, allowing us to increase the frequency with which we continually test how elements of the marketing plan combine to cause actions and reactions. Further, the access to Big Data presents opportunities to find more reliable proxies for the inevitable gaps we have in our “ideal” analysis set.
Cloud Computing has substantially reduced the cost of running simulations on models with dozens or even hundreds of floating variables, enabling a manager in minutes to assess a much broader and richer array of possible outcomes to an almost endless variety of tactical inputs.
And social networks provide the basis for early-indicator tools the likes of which never were never before available to marketers. Sure, they can be used as the technological “canary in the coal mine” to spot trouble emerging fast and switch plans to avert disaster. But more importantly, they can be used as early indicators of success – allowing us to know within days (sometimes hours) if a new program or initiative is likely to hit its goals or not – and then double-down or re-direct resources away from mediocre outcomes.
The answers to today’s questions regarding marketing’s impact on growth are hiding in the dark corners at the intersections of the new tactics and the old. Right now, today, there are more of these intersections than ever before. Perhaps an order of magnitude more. And the future isn’t likely to see the pace of growth slow much.
To find the answers, the science community needs to re-envision how to build analytics to capitalize on Big Data; how to evolve the mathematics to accommodate distributed computing power and calculate things consecutively vs. sequentially; and to re-think what “social networks” really are and how we use them. Those on the forefront of mathematics and computer science need to push the boundaries of machine learning in ways that complement human knowledge without seeking to replace it in threatening, “black box” ways.
To leap to the next level of insight and enable all this re-imagining, marketing scientists need to find ways to attract and concentrate investment capital. Research budgets have been shredded in the past decade as surveys first moved to the Internet and then social media. Marketing organizations are hiring analytics talent in small groups, but then over-whelming them with the day-to-day flow of ad-hoc reports to produce and backward-looking analysis as opposed to setting them on a mission to re-invent the fundamental methods required to bring credible answers to the most important questions. In short, the collective efforts are, like the tactical landscape we live in, increasingly fractured.
We need to find the next-level methodological foundation to more clearly connect marketing investments with growth. The good news is that the money is out there, searching for the clever, passionate, experienced minds possessing the vision to see marketing as more of a strategic growth driver than a “necessary expense”.