The quake that hit the global financial markets last year left shoppers and retailers bracing for a cold and lackluster Christmas. As anticipated, post-holiday consumer surveys reported holiday sales far below expectations as a result of record low spending. The only winners in the face of the economic downturn were value retail outlets such as Walmart which experienced increased consumer traffic (69% shoppers in 2008 as compared 33% in 2007; source: ARG), as a direct result of aggressive value-formoney promotions and discounts.
Dramatic shifts in consumer behavior, ranging from withholding spending to bargain hunting, have resulted in top line growth pressures for manufacturers and marketers alike. With budget cuts and increasing ROI as the mantra of the day, slashing marketing budgets is considered to be a given; as a result, marketers with limited budgets are obsessively looking at ways to justify marketing investments which can optimize revenue.
While many of us are busy dusting off the covers of our “how to survive in a shrinking marketplace” guide, savvy marketing gurus are turning to innovative solutions such as predictive analytics to improve ROI. For example, Walmart, Tesco, Home Depot, and Royal Bank of Canada, have turned to predictive analytics to produce tactical insights which translate into better targeted marketing strategies and resultant customer retention. CMOs are gradually waking up to the advantages offered by predictive analytics as compared to other decision-making tools in predicting, with near-perfect accuracy, the probability of a win versus loss of a sale under intensely competitive and disruptive economic conditions.
The low down on predictive analytics
Predictive analytics is a methodology that fundamentally draws knowledge from vast pools of data about individual customers. What makes predictive analytics compelling is the multi-dimensional approach it brings to understanding customer behavior. A variety of vectors such as demographics, buying behavior, discount hunting behavior, lifestyle preferences, income group, and transaction frequency are analyzed in a rigorous methodology to provide actionable insights.
This knowledge can then be used to “harness the hive.” Essentially predictive analytics helps the marketer sift through volumes of data, scrub and filter out the randomness to derive useable actionable information, deducing the underlying patterns which predict the relationship between consumer behavior and the drivers of that behavior.
Once deciphered, past consumer behavior patterns are used to develop models which predict future consumer behavior. The marketer is basically checking the rear view mirror to ensure unimpeded progress on the road ahead. Modeling, in effect, captures the subtle relationships between drivers of consumer behavior, exploiting them to tailor revenue-generating marketing strategies.
Let's leave aside discussions about serpentine equations and arcane statistical fundamentals. It is the patterns that they yield that matter; these patterns shape the development of predictive models deploying disciplined mathematical and statistical principles.
Predictive analytics can help marketers identify customers with a higher likelihood of responding to a particular marketing offer, in turn driving ROI. Targeting precisely these consumers will result in a greater response rate and help avoid expenditures on non-converts. But apart from more accurately identifying prospective customers, what other benefits can you derive by deploying predictive analytics?
By employing predictive analytics, such companies can learn how to optimally allocate a fixed marketing budget across multiple channels to maximize ROI. One of the best examples is the optimization of business-tobusiness (BtoB) marketing strategies. BtoB companies are usually multi-brand and multi-geographical with multimillion dollar budgets. Predictive analytics effectively allocates funding across multiple channels (such as direct mail, telemarketing, emails) to reach their clients with the highest return.
Predictive models help marketing teams understand and curb attrition by investing in and developing their customer retention programs. Banks, retailers, travel firms, and the hospitality industry commonly face “soft attrition,” i.e., customers slowly moving away from the business. Predictive analytics helps companies not only identify customers that are likely to churn but can also help identify drivers that will make them stay.
By analyzing product purchase sequences and associations, and characterizing them using predictive analytic modeling, it is possible, for any given product or service (be it groceries for families or the software/hardware needs of growing corporations), to exploit specific purchase relationships to cross-sell products of interest. Bottom line: greater ROI!
Predictive analytics can help such companies project volume forecasts, thereby optimize marketing mix spends and remove variance from production lines.
One of the world's largest brand-name apparel marketers with sales in more than 110 countries used predictive analytics to forecast unadjusted demand for a group of 93 product categories (PC’s) with a horizon of 9 months. As a result, the forecasting accuracy was increased by over 70 percent.
These are just a few of the applications of predictive analytics as a revenue optimization tool. Leading companies are aggressively augmenting their analytics capabilities by outsourcing their predictive analytics requirements to help develop a faster, sharper focus on customers - after all that's the best way to profitable growth.