Predictive analytics can help you use past data to predict likely future outcomes. While these techniques have long been used in forecasting, risk aversion, and even fraud detection, in the digital world, predictive can be extremely powerful.
These days the amount of digital marketing data and the availability of enormous processing power make the application of predictive analytics in the digital marketing space particularly exciting. It can be used to predict a likely lifetime customer value or the probability of customer loyalty or churn, and predictive techniques can also be used to gain insights into the best ways to allocate marketing budgets or understand the likely effectiveness of potential campaigns.
Check out this month’s Marketing Land column, What can predictive analytics REALLY do? Three case studies in seeing the future, which describes how a company in the auto aftermarket industry used predictive analytics to optimize digital spend towards actions that were highly predictive of future revenue; how U.S. Cellular received a 61% lift in visitor lifetime value by identifying which on-site actions were most likely to lead to purchases; and how a fashion retail group used predictive analytics to understand where to allocate their advertising budget and avoid spending money on sales that they would have acquired through organic means anyway.
Read the whole article on Marketing Land.
Want to learn more about predictive analytics? Hear more about how U.S. Cellular prioritized its website testing using insights from purchase intent analysis in this on-demand webinar featuring U.S. Cellular’s Digital Analytics Manager, Katie Birmingham.
Merchant:Southern States Cooperative
Market: 17 States
Size: 1,200 retail outlets
Platform: Alteryx, Facebook
Bottom Line: Predictive analytics tools help companies leverage transactional data and analyze trends to improve the performance of targeted marketing campaigns.
In Greg Bucko’s world, data is king. As the manager of customer insights for Southern States Cooperative, a farm supply and service cooperative with $2.5 billion in revenue, Bucko is responsible for making sure his company is targeting its most profitable customers with direct mail and other marketing calls to action. His company uses predictive analytics tools to analyze variations in marketing campaigns, with the goal of being able to improve customer targeting.
“We do a lot of ad hoc analysis regarding products people are buying, certain buying patterns, seasonality, unique SKUs of their purchases, regional differences, and weather effects. Those are all things that can be pulled through predictive analytics,” says Bucko.
Although Southern States’ marketing initiatives “cover the spectrum” — the company uses TV advertising, radio, print ads, direct mail, email marketing, and social media for outreach — Bucko says predictive analytics tools play an especially important role. The company has been gathering data through Alteryx, a strategic analytics and business intelligence vendor, for the past two-and-a-half years. “There’s not a day that goes by that we’re not using Alteryx within our marketing function,” says Bucko.
Bucko uses Alteryx to pull data from the “right list of customers,” and creates models of his customer lists to make sure he’s targeting the specific customers who are most likely to respond to direct mail pieces and other calls to action. “It’s such an important part of what we do when we’re analyzing trends of just doing basic ad hoc analysis, creating the visuals in order to understand the data better,” says Bucko.
Predictive analytics tools have also helped Southern States identify potential locations for new stores based on the rich connections to demographics and transactional data from current locations. “[The] ability to do mapping and to understand the context of a particular geography is incredibly important for a retailer like ourselves,” says Bucko.
Although Bucko had used business analytics tools previously, he says Alteryx has made it easier to pull data from multiple databases — such as his transactional databases and servers — with one piece of software. “I’m able to extract using Alteryx directly from our server. I’m able to scrub the data. I’m able to move the data into position where it can be modeled, then I’m able to remove the outliers and actually build a predictive model,” says Bucko.
Email marketing has proven to be more challenging than traditional direct mail, in part because Southern States doesn’t link email addresses to its transactional database. “In the future, though, we will be linking our email to our transactional data, so we can build the same kind of predictable models that we use currently in our traditional direct mail,” says Bucko.
Bucko also anticipates an increased push into social media platforms like Facebook, where Southern States already has more than 208,000 fans. “We’d love to be able to do more analysis and understand how well we’re doing with the customer experience, how well we’re meeting customers’ expectations, and using social media to leverage that,” says Bucko.
By using transactional data to identify buying patterns, regional differences, and weather effects, Southern States has been able to effectively target its direct mail pieces at those customers most likely to be responsive. However, Southern States is limited in the ways that it can use predictive analytics for marketing, since the company has yet to integrate email addresses or mobile phone numbers into its transactional databases. Once this integration occurs, Bucko will be able to take the same analysis he’s been using for his direct mail targeting and apply it to his email marketing efforts.
Stephanie Miles is an associate editor at Street Fight.
Click here to read more Street Fight case studies.