Trending Topics summarizes news, information and perspectives on matters affecting businesses and business leaders today. This insight is being provided to keep you up to date on the latest developments and trends influencing these topics. These views do not necessarily represent the views and opinions of PNC. For additional research on these topics, please consult the sources cited in this article.

Traditionally, companies trying to figure out their “next big thing" or new opportunities in the marketplace had to pore over sales history, seasonal trends, customer feedback, and other data to try to guess what their audiences want next. However, trying to use that data from disparate sources required no small amount of labor—with no guarantees that it would paint an accurate picture.

Today, powerful tools give businesses greater insight into what customers want. With the right technology, companies can use analytics to predict how customer needs and demands are changing and adapt accordingly.

“The value of predictive analytics can be realized across the entire customer journey—to engage prospects and customers as they discover a need or address a pain point, to power cross-sell and up-sell, reduce churn and customer attrition; to grow lifetime account value; and even convert loyal customers into brand ambassadors," says Wilson Raj, global director of customer intelligence at SAS in Cary, North Carolina.

Start With Good Data

First, you need to start with a good body of real-time and historical data, says Tim Mehta, senior optimization strategist at Portent, Inc., a Seattle-based marketing agency owned by Clearlink. “The applications that predictive analytics provide for small to midsize businesses (SMBs) are highly dependent upon the sample size of the data that's being collected. If an SMB only gets a few hundred visits a day to their site or application, the time to accumulate enough data to be statistically significant would be too long to keep up with the pace that the SMB market moves," he says.

For those who have a greater body of data—more customers, and a greater number of interaction points— predictive analytics works much like it does at enterprise companies. Say you are a retailer with physical locations and an e-commerce website. Mehta says you can make fundamental assumptions about factors that might influence a user's behavior, such as age, device, location, how many times they've visited the site, how many times they've seen an ad, and so forth. You can then use predictive analytics to look for concrete patterns tied to these factors so you can tailor your content to best suit your users' needs.

Various apps and platforms can analyze customer behavior and previous buying patterns and, from that, make projections about what customers want as well as what the business needs. These may include:

  • Suggested re-order prompts
  • Recommended products based on similar purchase patterns
  • Inventory and re-order prompts when stock gets low
  • Lead scoring based on customer profiles
  • Seasonal buying patterns
  • Personalized promotions

Predictive analytics based on a good, comprehensive body of data can help you increase brand engagement and deepen loyalty, Raj says. He points to one company which used predictive analytics to personalize offers. The efforts worked so well that 25 percent of the solicited customers responded to the offer and 14 percent accepted it. That compares with a rate of 1 percent before the company began using predictive analytics for personalization, he says.

Applying Predictive Analytics

Companies that are interested in applying predictive analytics can look to tools like SAS, InsightSquared, Google Analytics, and Canopy Labs that enable you to analyze everything from past buying behavior to inventory to sales trends. Look for tools that integrate with the platforms you use and the data you have from your customer relationship management (CRM) system, website, mobile, social media channels, etc.

Next, map the right data sources—and, most important, the analytics—to specific stages of the customer journey, Raj says. “It's about accurately anticipating your customers and designing specific initiatives to put those predictive insights into action," he says.

For example, during the purchase phase, predictive analytics can help businesses understand how and when customers will purchase. Predictive techniques such as propensity models help marketers predict the likelihood that a customer will respond to a specific offer or message and convert. They can also help you cross-sell or identify other products the customer is likely to purchase, helping you boost transactions or target customers accurately with in-market timing models, he says.

Predictive analytics can also help uncover patterns of usage behavior and further drive customer engagement. For example, a retail site may tell you the status of your recent order the moment you land on the home page. “Churn models" can help you spot signs that a customer is about to abandon your business or shop somewhere else, giving you the opportunity to re-engage them, such as special offers or free upgrades.

Also, pay attention to privacy. With regulations such as Europe's General Data Protection Regulations (GDPR), companies can leave themselves open to liability if they don't consider proper privacy protections.

“There's something spooky about a brand 'knowing' something about a consumer's innermost motivations and preferences. Yet, consumers expect relevant, well-timed personalized communication and offers. Surround your predictive analytics efforts with thoughtful data stewardship efforts," he says. “Better still, augment your predictive analytics approaches with good old-fashioned human engagement that uniquely defines your brand."

Ready to Help

At PNC, we combine a wider range of financial resources with a deeper understanding of your business to help you achieve your goals. To learn more please contact your Relationship Manager or visit pnc.com/cib.