|
|
Media Coverage
"Full
Speed Ahead: Real-time Decision Management Solutions Help Advertisers
Make Changes to Campaigns on the Fly"
The Advertiser
December,
2007
From cars that parallel park themselves to vacuum cleaners that search out dirt on their own, automation is making the world easier to navitgate. But as most marketers would undoubtedly agree, real-time automated business processes, an innovation that has changed the way most companies operate, have been slow to transform the advertising and marketing industry. That about to change. Thanks to a wave of real-time decision management solutions, marketers now have the ability to not only streamline everything form project management to media buying but also change directions at a moment's notice.
Although the emerging field of real-time decision making is an obstacle course in provincial terminology, varied in its objectives and largely unproven in its results, the upside is that it offers marketers the promise of in-depth, split-second planning and reporting. "It's a very good ROI methodology," says Arjen Linders, vice president of marketing for New York-based Philips Norelco, who began utilizing the technology several years ago. "most important for me personally is that it's very timely - I have a real-time, online feeling with this. So during my hard six weeks from Black Friday and Christmas, I can really change my own destiny."
What is Real-Time Decision Making?
According to Carol-Ann Matignon, vice president, product management, enterprise decision technologies, at Minneapolis-based Fair Isaac and an evangelist for the process of automated decision making, three kinds of technologies are applied within the realm of real-time decision management solutions:
- Simple code, which executes certain business functions. Code is prevalent because most businesses already have applications; they just need them to be smarter.
- A declarative rules-based approach that says if a certain requirement is met, a certain action occurs.
- Analytics, or tolls, that generate executable code from the results of analyzing data and events. It can be used, for example, to predict how likely a customer is to take a specific action based on existing data.
"The biggest energy at the moment is in analytics," Matignon says, noting that a lot of work has been done in predicting purchase patterns, customer loyalty, and offera acceptance. This is the fastest growing area."
While the use of predictive modeling for real-time decision making is quickly growing in marketing and advertising, it's not necessarily new technology in the area of risk management. For over 50 years, Fair Isaac has used the technology for credit scoring and credit fraud. "Now the company is adapting it to more opportunity-based industries, like advertising and marketing, automating everything from CRM to media planning.
The marketplace for these services is, of course, already crowded. For example, Cary, N.C.-based SAS has a real-time decision solution that allows business users to guide interactions with a customer using business rules, process management, and predictive models. New York-based [x+1] is a predictive marketing firm that allows marketers to make online advertising investment decisions, also in a real-time environment. The company leverages its patented predictive optimization engine that analyzes all types of data. Oaks, Penn.-based MediGroup has a program called Silent Partner that specializes in project management decisions; the company has been refining its product for the decision management market over the past three years.
Almost as challenging as understanding what comprises real-time decision making is being able to spot the technology in the wild. Every company that offers a real-time decision-making solution has its own terminology for it, and as the industry takes shape, each is fighting to make its own words stick. Predictive modeling, business performance management, and project portfolio management are some of the terms bandied about.
Then there's the dust-up between companies that look at historical or descriptive data rather than predictive analytics. "Under the umbrella of predictive analytics, there are actually some kinds of analytics that are not predictive - they're descriptive - but descriptive analytics are really what people think of in terms of reporting and dashboards," Matignon says. "We are struggling as an industry for a phrase for this."
What Can Real-Time Decision Making Do?
In the 1990s, Philips Norelco had a problem. "Every year we took out more advertising money, and we saw our sales and share of market decrease," Norelco's Linders said at the 2007 ANA Marketing Accountability Conference. Worldwide, Philips is the leader in electric shaving, but in the past decade, blade shaving has stepped up considerably. So Philips Norelco tapped El Dorado Hills, Calif.-based Marketing Evolution to help the brand boost market share and launch a new line of body shavers. Marketing Evolution's solution, called ROMO (Return on Marketing Objective), grew out of its work with blue-chip advertising companies and monitoring more than $1 billion in advertising revenue. The company begins with extensive scenario akin to the "war games" used by the government to calculate military contingencies. Cliches and hyperbole aside, the "business is like war" analogy is understandable considering that some of the budgets rolled into the big-ticket campaigns.
Rex Briggs, the CEO of Marketing Evolution and coauthor of What Sticks? Why Most Advertising Fails and How to Guarantee Yours Succeeds, sees the need to plan out all eventualities because oftentimes marketers don't control all of the variables. "It's not like a factory process where you program in the decisions that will be made," he says. "You never know when you launch a campaign what your competition will do, for example."
For Norelco, however, Marketing Evolution mapped out the perfect scenario. With a marketing budget of only $900,000, Philips Norelco wanted to sell 150,000 of its new body shavers. The campaign was aimed at a younger demographic than its 35-plus sweet spot, and considering the low budget, the company decided for the first time to try a viral campaign. It set clear KPIs to monitor the campaign, like 100,000 weekly Web site hits and 10 percent awareness in body grooming.
Within six weeks of the launch, the campaign was clearly a hit. Norelco was using the Web site, which it announced on the Howard Stern radio show, to monitor the buzz using strategy and scenario planning. The Wall Street Journal and The New York Times picked up stories on their Web sites, and Norelco got 100,000 hits per day over the first four days. At the end of one year, it had 3 million visitors to the site, more than Phillips.com receives in a year.
What's more, the company beat its sales goals by 100 percent. "Because of the funnel we created, we were able to get extra budget, and without any financial risk, we were able to grow faster," Linders says. Norelco invested $150,000 more because it was seeing great results, which ended up netting the company twice as much in sales as its goal. "You can steer while the campaign is running," he says, "and we n ever had that feeling with our previous campaigns."
What Are the Downsides of Real-Time Decision Making?
At its worst, Andy Bober, director of customer intelligence product management at SAS, likens real-time decision making to a recent film in which police were able to detect crimes in advance of their occurring.
Because of the service's ability to predict customer behaviors, it also has the ability to creep the consumer out. "While you might be able to make an offer right away, you need to be respectful," Bober says. "Just because you can, doesn't mean you should."
Then again, the creep factor is all relative. For example, 1800Flowers.com will begin implementing SAS' system over the holidays. The Carle Place, N.Y.-based company is using it primarily to augment its Web site and call center CRM, even though SAS also offers a marketing resource management solution called Veridiem MRM that allows advertisers to analyze the overall spend.
"It's about using the knowledge we already have about our customers and making their experience more relevant," says Aaron Cano, vice president of customer knowledge for 1800Flowers.com. "The customers have come to expect us to know who they are. If I'm shopping for my wife's birthday, 1800Flowers.com should know that I always buy my wife yellow roses, and if we have a special they should make it available to me."
Another potential problem with real-time decision management is likely to be that which draws the most interest in his phenomenal technology. While it may be tempting to "set it and forget it," it's more about business at hyperspeed, not on autopilot. "It's not really the real-time data that matters," Briggs says. "It's the fast reaction of marketers to that data that makes the difference."
<< Return to Previous Page |
Privacy
Site Map
Home
![[x+1]](../images/logo_xplus_main.gif){kind=logo}
