shopify stats

Using Customer-Centric Predictive Analytics


Predictive analytics is about the consumer, and bringing a holistic view of that consumer. To do that you have to understand the profile of the consumer. They are more than a segment,” explains Emad Georgy, SVP of product development and global head of development at Experian Marketing Services. “Technology has caught up with that concept. Now with big data, we have technology that’s bringing true predictive analytics to the table.”

Online and mobile technologies are expanding rapidly, and they are impacting customer expectations. One of the most important ways organizations are meeting these new customer expectations is by taking advantage of the power big data and analytics offer. For example, there are now many data sources available using social media platforms and websites. Predictive analytics anticipates future consumer behaviors based on large amounts of current and past indicative data that has been collected from multiple sources. Marketers now realize that customer experience is part of the competitive advantage, and predictive analytics is a tool to aid in better understanding their customers. Predictive analytics is way to employ techniques, such as modeling or data mining, to gather information from data and apply the information to predict future trends and behavior. It can analyze current and/or historical data to make reliable predictions about the future behaviors or events.

Predictive analytics optimizes marketing campaigns and website behavior to increase customer responses, conversions and clicks, and to decrease churn. Each customer’s predictive score informs actions to be taken with that customer — business intelligence just doesn’t get more actionable than that.

Although predictive analytics may seem very new, we actually would have a hard time getting through a day without encountering predictive analytics output:

  • The web ads that show up as we are navigating the internet are targeted for us based on predictive models
  • Our credit score is developed using on a predictive model that analyzes our past behavior and existing liens
  • Our medical insurance company has assigned a risk profile to us based on our medical history and lifestyle information
  • When we are selecting a movie to watch through our online provider, suggestions are made using a predictive model that analyzes our previous viewing history and likes
  • When we use our loyalty cards to get coupons or special offers, we are getting targeted coupons based on our previous sales history.

Predictive analytics is typically a four step-by-step process:

  • Establish objectives – Establish what you want to achieve, develop hypothesis with experts and the data required.
  • Collect good quality data – Establish a view of customer combining enterprise and social media opinions, intent and sentiment including unstructured and free form data such as comments, emails, tweets, Facebook posts and SMS messages.
  • Understand behavior and intent – Understand customer’s behavior and intent across channels and platforms by deploying predictive analytics in conjunction with organizational wisdom.
  • Predict action – Predict a customer’s next purchase and make the right offer at the right time and in the right way.  Evaluate and adjust as required.

LinkedIn’s Head of Marketing Products, Russell Glass, made an observation regarding the importance of using customer-centric data effectively during his keynote address at DMA’s 2015 Marketing Analytics Conference. He said, “Becoming customer-centric is the backbone to using data effectively. I go on Amazon and I have a completely different experience than you.  They’ve created a very customer-centric experience…They’ve put businesses like Borders — who were not customer-centric and who thought people would always want to come in and touch books — out of business.” He continued, “What is an amazing experience for your customers that you can create?” he asked.  “The Four Seasons — they know me, they have the bottle of wine I like, know my preferences. That engenders trust,” he said.  “You have to ask yourself ‘What do I know about my customer? What do I want to know about my customer?’”