Decisions, Decisions

Article Featured Image

For the rest of the May 2009 issue of CRM magazine, please click here.

Dum dum da-dum.

Wedding photography doesn’t immediately strike a chord as being a forward-looking, tech-savvy industry, does it? Well, at San Francisco–based Bella Pictures, which matches regional photographers with clients based on their styles and preferences, operations involve more than capturing smiling faces. The business model is complex—far beyond a bride merely calling up and booking a photographer. Bella Pictures often pops the question on its own, snapping up prospects through name-capture promotions such as sweepstakes.

In a presentation at the recent Predictive Analytics World conference, Traci Chu, Bella’s director of CRM, recalled a realization: The market was changing, and Bella’s lead-scoring method was not. The scoring system in place was effective in prioritizing calls to make, and it shared with representatives a number of important factors that influence a buyer’s decision—the wedding date, the budget, the venue, the time until the wedding, and whether the lead had filled out a Web-site form. (For more on influencing sales, see “The Psychology of the Sale.”) However, the system neglected to account for any previous interest shown by leads, such as whether they received email offers or if a bad address had led to a bounce. “Marketers’ intuition doesn’t always include negative variables,” Chu said. “So why keep investing in a lead if the messages aren’t getting through?”

Bella Pictures, after assessing all the factors not included in its process, turned to the expertise of Business Researchers to deploy predictive analytics modeling and to start segmenting prospects appropriately. Chu says leads are now scored in real time, rather than just after a potential customer submits a form. The predictive models include both positive and negative variables that are weighted using statistics—complex stuff for a bunch of wedding photographers. And yet, although the implementation has just begun, Chu says she expects to save money now that she can target just the most-interested prospects, and her agents can make the right decisions at the right time. Because, in the end, it all comes down to decisions.

Basic Decisioning
Bella Pictures calls itself a decision-centric company, having reorganized its business operations to make the decision process more accurate. Decision centricity in no way trumps customer centricity—the two are actually complementary. The focus on decision making has even earned its own umbrella term—enterprise decision management, or EDM. Before jumping ahead, though, it’s important to note that EDM—like CRM—is about more than just technology.

If you were under the impression that the technology alone was the point, James Taylor, who was among the first to use the term EDM, would likely scold you for it. Taylor, author of the book Smart (enough) Systems, has written extensively on the topic of EDM and says decision management is more of a philosophy or culture. It’s about competing on decisions, and the software that enables an organization to do so is only part of the battle. Technologies such as price management, real-time decisioning, predictive analytics, text and data mining—they sit underneath the umbrella, but on their own they can’t protect your company from inclement weather.

Taylor says that the EDM term might trip up the uninitiated. “The ‘enterprise’ word makes companies think they have to do it for the whole enterprise,” he says. “The reason behind the ‘E’ was not so much that you have to [make] every decision enterprisewide, but…you have to make [every decision] into an enterprise asset,” Taylor explains.

EDM also spans silos. More than just a marketing strategy, the methodology is relevant for all parts of the enterprise. The idea of competing on decisions is spreading. In fact, the Predictive Analytics World conference Bella Pictures’ Traci Chu presented at was the first of its kind. Taylor, who also presented at the show, equates the discovery of EDM to that of business process management (BPM). “Before BPM, people always wanted to have efficiency and processing, but they struggled to make it happen because they didn’t have a framework,” he says.

As you might expect, EDM often boils down to making the right offer for the right person in the right way at the right time. But how exactly does an organization start attacking EDM and making its decisions meaningful? By thinking about business at a high level and asking questions: Which interactions with customers are most critical? Is there an area in which efficiency is lacking? When is proper timing vital?

Sound complicated? Yes—perhaps that’s why Eric Siegel, the conference chairman at Predictive Analytics World, called attendees “pioneers.” Adoption has been hard to come by. The topic is complicated and users are still grappling with it even at the most basic levels.

“Many organizations still struggle with really understanding the value of analytics and what [those figures] are telling them,” says Kathy Konkel, manager of product marketing for SPSS, a provider of predictive analytics software. “We are focusing on trying to communicate to business users what the model is telling them and how they are able to quantify that. It’s viewed as this ‘black magic’ stuff.”

Is EDM Really So Complicated?
Experts, however, firmly believe that EDM’s far less mystical than it seems.

Consider the science behind automatic car windows: What’s powering the up-and-down motion? Freescale, a company in Austin, Texas, that manufactures microcontrollers and semicontrollers, could probably tell you—albeit in a highly technical manner. Freescale competes in a complex field—complexities that involve not just the products it sells, but the manner in which they’re sold.

“There’s huge lead time in the automotive industry,” explains Kimberly Appleton, the company’s corporate pricing manager. “When we quote something, we might not sell the product for the next couple of years.” Because the selling process can be so drawn out, effective quoting is imperative. In 2006, the company turned to Vendavo, a pricing management provider, for help with pricing decisions. “With a portfolio as large as ours, we do a lot of custom parts designed specifically for a customer,” Appleton says. “We also do a lot of standard parts, which allows us to do analysis of profitability as well as [of] pure bottom-line revenue at the same time as focusing on our R&D efforts.” With intelligence and automation behind its pricing process, Freescale saves both time and money, and is able to compete globally without misquoting.

All too often, a company makes major decisions simply according to the way previous decisions were made. Employees, by nature, bring a bias to businesses. Marketers and salespeople rely on intuition and assumptions based on how they’ve done things in the past.

Colin Shearer, senior vice president of strategic analytics for SPSS, says that today’s CRM provides limited benefits in terms of decision-making prowess. “It’s making interactions happen and making sure they’re recorded and efficient,” he says. Unfortunately, a typical CRM system “is not making sure they’re the right interactions.”

The Changing Climate
Bella Pictures has come to terms with the idea that, despite all the hard work put into creating its five models for lead scoring, those models will only last a year at best. Businesses and customers change at the speed of light—and analytics should reflect that, not contradict it. Taylor puts it bluntly: “Doing something the same way and expecting a different result is another definition of insanity.” He says that the challenge is not in building the models—or in choosing or implementing the technology, for that matter. The challenge is in putting those models and that technology into place to change how people do things and how decisions are made.

“Too many models are built into a process that assumes once the model is complete, it’s done,” Taylor says.

New DVD releases, for example, hit store shelves each week at 12:01 a.m. on Tuesdays, as regular as clockwork. And yet that level of predictability masks an opportunity for a predictive model regarding how those DVDs are displayed, says Jay Venkateswaran, senior vice president of research and analytics for WNS Global Services, an outsourcer for predictive analytics solutions.

WNS, which can replace a company’s need for an in-house statistician with an overseas one of its own, hosts a solution for a major retailer that has created models for the chain’s DVD distribution. Venkateswaran says about 70 percent of all DVD sales occur within the first week of release, meaning an inconsistent and unintelligent ordering system can lead to over- or understocking—and missed revenue opportunities.
WNS implemented a predictive modeling solution to ease the retailer’s decisioning—and now determines what movies are sent to what stores and in what quantities. Many factors go into the decision process, including geographic location, ratio of male shoppers to female ones, the number of mothers who shop at the store, the proximity to schools, etc. According to WNS, the forecasting implementation now saves the retailer $3 million annually—a number that speaks to the power of the prediction.

Take it Personally
“People respond to the way you treat them as if it was selected personally and deliberately for them,” Taylor suggests. “They can’t help it. That’s how people are.” Think about how you react to a piece of junk mail or an unsolicited phone call. Despite the potential for visceral impact, Taylor admits, most customer-treatment decisions are impersonal, and in many cases accidental.

But what if marketers treated each letter or marketing message they sent as a decision? “Let’s actually mean to send them this letter,” Taylor proposes. The results are phenomenal, he says. In fact, one company he knows of got intensely personal with its targeted newsletters—and saw a 2,000 percent improvement in response rate. If that’s the case, why isn’t personalization happening across the board? Taylor says it’s because organizations aren’t thinking about the decisions as the differentiator—and they simply aren’t asking the right questions.

Competing on Decisions
“The big difference now is organizations are waking up to the potential of putting the resulting decisions [on] the front line,” Shearer says. “That’s what’s transforming decisions.”

This decision management stuff isn’t just for the tech-savvy corporate bigwigs. It’s for employees running the phone lines and responding to email inquiries at Bella Pictures. It’s for the Coca-Cola stockperson who strategically lines the grocery shelves alongside Pepsi products. Michael Porter, Harvard Business School professor and authority on competitive strategy, sums it up nicely: “The essence of strategy is choosing to perform activities differently than rivals do.”

Bella Pictures’ Chu says that, in the next modeling effort, she’d like to integrate such factors as call patterns and historical data, interest in high-value content pages, geography, and video interest. Doing so will make Bella’s marketing efforts more personal.

The organizers of Predictive Analytics World collected survey results from attendees who claimed experience with predictive analytics solutions. Even given the self-selecting sample set, the results speak to the momentum around this topic and technology: Ninety percent of respondents who have deployed predictive analytics said that their most successful initiative had achieved a positive return on investment.
No one said making a decision was easy. “We tell [customers] predictive analytics isn’t a ‘Big Bang’ approach,” Shearer says. “It’s a journey.”


SIDEBAR: Defining Decisions
The Data Warehousing Institute (TDWI) defines data mining as the “discovery and modeling of hidden patterns in large volumes of data.” Data mining can discover those patterns and build models automatically, without looking for something specific. Unlike such retrospective (backward-looking) technologies, predictive analytics produces forward-looking models that capture and represent patterns and interactions.

Vladimir Stojanovski, a blogger and an engagement manager/solutions architect, offers a nice metaphor to describe the relationship between predictive analytics and business intelligence (BI): “If BI is a look in the rearview mirror,” he writes on his blog, “predictive analytics is the view out the windshield.”

EDM goes even further. “Data mining,” writes Carole-Ann Matignon (EDM Blog), “shares many useful techniques with EDM but EDM is more than just data mining on steroids.”

Here’s a quick snapshot of the progression from data mining to predictive analytics to enterprise decision management:

Data Mining In simple terms, the analysis of data for relationships that have not previously been discovered. Although often used synonymously with the phrase “predictive analytics,” data mining involves sifting through very large amounts of data for useful information. Using artificial intelligence techniques, neural networks, and advanced statistical tools (such as cluster analysis) to reveal trends, patterns, and relationships that might otherwise have remained undetected, data mining attempts to discover hidden rules underlying the data. It can also be called data surfing. (Source: www.businessdictionary.com)

Predictive Analytics An activity that allows us to quantify future events or actions. This quantification could be as straightforward as generating a list of customers who are likely, at a point in time, to behave in a certain way (e.g. to churn, to register, to buy, or to respond to a particular mail). Typically this list will be accompanied by a score that provides the probability (or propensity) that an event will occur. The forecasts may alternatively be presented as confidence values. (Source: Foviance)

Enterprise Decision Management An approach for automating and improving high-volume operational decisions, EDM develops decision services using business rules to automate those decisions, adds analytic insight to these services using predictive analytics, and allows for the ongoing improvement of decision making through adaptive control and optimization. (Source: James Taylor, author, Smart (enough) Systems)


SIDEBAR: Predict Your Way to Savings
Five ways to cut costs using predictive modeling:

  1. Don’t contact those who won’t respond.
    Response modeling—This type of modeling basically means targeting fewer people without sacrificing a high response rate. By modeling on top of historical customer data, you can see who is likely to respond to an offer, and then target accordingly.
  2. Don’t contact those who would have made a purchase anyway.
    Response-uplift modeling—The next step after response modeling involves segmenting out those who would likely purchase regardless of an offer. Having identified this segment, you can save money by removing them from the campaign.
  3. Don’t waste expensive retention offers on those who will stay anyway.
    Churn modeling—Retaining customers can be quite expensive. You can’t afford to offer [retention incentives] to all of your customers. If you predict those most likely to leave, you can target much more efficiently.
  4. Don’t offend those who would otherwise stay.
    Churn-uplift modeling—The cliché is true: It’s sometimes best to let sleeping dogs lie. Some customers would never have taken their money elsewhere until you reminded them you had it. It’s better just to leave these stagnant customers alone. You don’t want to remind them that they’re paying you, or annoy them with unnecessary offers. In order to do this, you have to model a control set of customers to properly segment—and measure the value of modeling initiatives.
  5. Don’t acquire “loss customers.”
    Risk modeling—You wouldn’t lend your credit card to a convicted thief, would you? By the same token, you don’t want to sell to customers who lack the means or the intention to pay. Risk modeling is especially crucial for credit-card providers and insurance companies, where if things go wrong the provider eats the cost.

Source: Eric Siegel, Predictive Analytics World 2009

Assistant Editor Lauren McKay can be reached at lmckay@destinationCRM.com.

Every month, CRM magazine covers the customer relationship management industry and beyond. To subscribe, please visit http://www.destinationcrm.com/subscribe/.

CRM Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues

Related Articles

Forrester Gives a Welcoming Wave to Complex Event Processing

Citing heavy demand, the research firm unveiled its first-ever evaluation of nine vendors in the rapidly maturing space.

Analytics in a Multichannel World

Make interactions more sophisticated.

Big Red Roundup

Oracle releases financial results, announces the departure of an influential leader, and creates news around its Fusion Middleware Apps.

The Psychology of the Sale

There's a lot going on inside the customer's head, whether you put it there or not. What are salespeople up against?

Buyer's Guide Companies Mentioned