The usage of data, analytics, and algorithms is central to both the current and future state of marketing — but many organizations still struggle to keep their marketing and data analytics teams integrated and on the same page.
In 2015, research conducted by Forbes Insights found that a mere 22 percent of large-scale data-driven marketing initiatives were “achieving significant results.” Unsurprisingly, a more recent Gartner study found that, “through 2017, 60 percent of big data projects will fail to go beyond piloting and experimentation, and will be abandoned.”
While these figures may have improved over the last three years, data-driven marketing efforts still seldom deliver on their full potential — at least not in fields like healthcare marketing. Insufficient data literacy among marketers (particularly those in more traditional, non-digitally native industries) is partially to blame, but an ongoing organizational disconnect between data analytics teams and marketing teams also poses consistent barriers to success. When data is mishandled, misunderstood, or presented in the improper context, key messaging — and a brand’s reputation — hang in the balance.
Of course, marketers rarely compromise the integrity of data analyses intentionally; there are countless things they must take into consideration in determining what to communicate to consumers, and data teams themselves aren’t always able to hand off their findings effectively. The end result, however, is that many data-driven marketing initiatives end up as haphazard assemblages evocative less of Dr. Frankenstein’s genius than his monster.
An All or Nothing Proposition
In practice, this can cause a variety of problems, as consistency and integration — connecting the dots, so to speak — are critical to any effective data-driven marketing campaign. “The integration of the dots is more important than the dots themselves,” argues Yale Professor of Management and Marketing Dr. Ravi Dhar. The key for marketers, he continues, is to “understand what they see in social media, what customers bought in a store, and what other media did they see to get the overall understanding of how media consumption drives both online and offline purchases.”
This kind of “layered listening” is central to what we do at Saatchi & Saatchi Wellness (SSW), as many of the actions we are trying to influence in health and wellness, given that almost all key decisions take place offline. In my role as Data Scientist at SSW, I lead many of our predictive customer segmentation and targeting engagements. Our team gathers data on hundreds, if not thousands of specific customer behaviors to train algorithms designed to give us a peek into who these people are and how they make decisions.
Once these analyses are complete, we create specific customer micro-segments based on predicted behaviors — one group of potential customers may be likely to try a new product based on a certain set of triggers, while another group of historically loyal customers may be likely to discontinue usage based on a different set of triggers. Ultimately, my job is to empower marketers with a set of tools that they can use to precisely target only the customers based on specific strategic intersections of observed and predicted behaviors.
When marketing teams do embrace analytics teams’ research whole cloth — and when data teams effectively communicate the importance of doing so — the results tend to be favorable. After all, human behavior is complex, and so the problem with taking bits and pieces from data analyses is that it can be difficult — and, if machine learning is used extensively, even impossible — to understand the numerous causal relationship between various “dots.” Social media may have a direct but difficult-to-see influence on, say, the effectiveness of a company’s paid search advertising. This influence will be embedded deep within an analytical model, but if a marketer chooses to focus only on paid search analytics and not the entire comprehensive model, they will end up neglecting critical touchpoints at the top of the sales funnel.
Forging Strong Marketing-Analytics Partnerships
Bridging this disconnect will first and foremost require stronger, more trusting marketing-analytics partnerships, and this begins with securing buy-in from all parties involved to drive organizational transformation. “The first step is a cultural shift,” confirms LinkedIn Head of B2B Product Russell Glass. “It starts at the top, in the leadership of the company at large…They need to acknowledge, ‘We are going to start to make decisions based on data.’”
This is most effective when an organization’s analytics and marketing teams are co-located, part of the same cross-functional team, and operate under one umbrella. When a marketer can walk down the hall and run through an analysis they don’t fully understand with the person who conducted it, the likelihood that they actually take action skyrockets.
That said, with good — read: consistent and transparent — communication and a bit of patience, any analytics and marketing team can forge a strong enough relationship to make data-driven marketing work. As data gets bigger and digital touchpoints become an increasingly consequential part of the marketing equation, this will become less of a option and more of a requirement.