neon sign advertising a psychic in a building's window
PHOTO: Jose Antonio Gallego Vázquez

Many marketers look at predictive analytics, machine learning (ML) and artificial intelligence (AI) like they do the Jetsons: as a future where an AI-based robot Rosie intelligently takes care of tasks while the people worry about working a three-hour week. The reality, however, is much different. As predictive analytics plays a growing role in the marketing suite, a very pragmatic approach informs how marketers can apply predictive analytics to hone their work today.

While ML and AI may in their final stages automatically take actions for marketers based on data inputs and predictive algorithms, that is not the case today. Marketers today have the opportunity to rethink their expectations of these technologies and how they can help make us more effective in lead generation, conversion and building long-term customer value by balancing the smarts of predictive technologies with the best of marketer expertise.

The customer journey is the best place to start resetting expectations around predictive analytics, which I define as an umbrella term for using a bunch of techniques to analyze current data to make future predictions. From modeling and validating to measuring and optimizing, predictive analytics in the customer journey can help make us all more productive marketers.

When Predictive Analytics Meets the Customer Journeys

Modeling

The customer journey is no longer a linear path. In fact, any one person might take any of a myriad of paths. While these paths may seem random and much like the path of a disc in a plinko game, predictive analytics can help us become smarter marketers by connecting digital breadcrumbs that aren’t readily apparent otherwise.

With individualized modeling powered by data insights, marketers can target both at the segment and individual level. They can optimize campaigns, content and more for prospects as they move along the path to purchase — with technology that allows them to do so at scale.

Validating

Validating customer journeys has traditionally been an intensive, manual exercise to understand the key campaign elements, like lead prioritization, nurturing, content and more, that provide better insights and validate what is the best course of action for a marketer to take at a given point in the customer journey.

Applying predictive analytics allows us to more quickly and efficiently confirm that we are capturing the steps that prospects take, identifying important milestones in the journey that keep them moving forward and flagging previously unknown gaps. Capturing data points in real-time validates what’s the best content, nurturing, or other step to take, and in some cases will make recommendations or potentially automatically take action to deliver the right touchpoint to the right person at just the right time.

Measuring

Measurement has traditionally offered a lagging view of success. Yet, ML and predictive analytics can give us a more real-time measurement of what’s working and what’s not. As algorithms are able to take a wide variety of data inputs, quickly correlating massive amounts of data, marketers can within moments have insight into metrics like lead velocity that help us not just automate, but optimize our processes.

Optimizing

People prefer to do business with companies who make their experience more relevant at every turn. Predictive analytics can help achieve this by not just personalizing but individualizing the customer journey. Smart marketers apply predictive analytics insights to go beyond mere automation to create a customer journey optimized for individual prospects and specific business outcomes.

For example, marketers at an electronic component manufacturer took customers’ digital breadcrumbs and combined them with measures of what was working to optimize customer touchpoints. By tailoring content to different prospects at different times in the buying process, the marketers successfully increased its conversion rate by over 1000 percent.

Related Article: Personalization Isn't the Goal, Conversion Is

When to Apply Predictive Analytics

When to use — and just as importantly, when not to use — predictive analytics is important when balancing automation and marketing expertise. Let’s quickly assess three areas that play a role in optimizing the customer journey.

  1. Content — Predictive analytics can validate what content performs best across measures, such as among certain market segments, and/or at certain stages of the buying journey. More than just measuring content, predictive analytics can auto-recommend content that will best help the prospect move forward in the process, and can even automatically take action to serve content to a specific prospect at the time they will be most receptive to it. These data-driven actions are constantly assessed and updated for constant improvement.
  2. Nurturing — Predictive analytics is helpful to honing marketer effectiveness by illustrating what nurturing activities work best for which market segments. For your defined segments — whether they be people living in a specific geography, or more nuanced — predictive technologies  can inform the best way to nurture a given segment. These data-driven systems will update your nurture flows to create a virtuous cycle of nurture paths that drive greater and greater value to your organization, and to customers.
  3. Leads — Marketers can expect predictive analytics to help float top targets to the top, with the combined use of prospect profiles and behaviors. Predictive analytics shines when it comes to analyzing mounds of prospect behavior, and it can quickly connect the dots between valuable activities and those touchpoints with lower influence, and recommend which activities lead to faster conversions and are therefore of higher value. Predictive analytics is "always-on," constantly collecting new data points that can help marketers develop a process of continuous updates to lead scoring, getting prospects in front of the sales team at the optimum time.

At the end of the day, customers are interacting with brands in more ways, through more channels, than ever before. As the customer journey becomes less linear and more individualized, there is a clear opportunity to pragmatically apply predictive analytics, ML and AI to make sense from the noise. Marrying the smarts of predictive technologies with the best of marketer expertise, smart marketers can create competitive advantage through modeling and validating, measuring and optimizing the customer journey.

Related Article: Bridging the Gap Between Data Analytics and Personalization