CRM News: Customer Data: Getting Creepy With Your Data
By Erika Morphy
“Some of Microsoft’s latest moves have a distinct ‘Big Brother’ feel,” wrote blogger Ann All. “There’s its application for a patent on a system that would allow managers to monitor workers’ physiological states through their computers — ostensibly to offer help to folks when they become frustrated. And then there’s the partnership with MediaCart Holdings that promises to serve up advertisements geared to the position of a shopper’s grocery cart in the store.”
Customers’ privacy and their data are two immutable and opposing forces that a company must always try to balance, usually unsuccessfully. Few for-profit firms have proven themselves able to resist the lure of monetizing their customers’ and users’ data.
Some consumer advocates have put up a good fight in keeping such privacy intrusions to a minimum. Still, though, laws outlining the proper use of customer data to protect against identity theft as well as to protect against predators are still, in many cases, stalled in the legislative process, says Kathryn C. Montgomery, a professor at American University 
and author of a chapter in the MacArthur Series on Digital Media and Learning: Civic Engagement. That is why a public debate on these issues are always welcome, she told CRM Buyer.
Meanwhile, however, it seems as though companies are one step ahead of lawmakers and even advocacy hawks, developing programs using customer data that go far beyond what a customer ever envisioned when she agreed to do business with the company.
Microsoft Gets Creepy
One recent blog by Ann All, that she titled the “Visible Enterprise,” describes how some companies are crossing the border from cool to creepy.
She singles out Microsoft (Nasdaq: MSFT) 
— although giving a nod to Google’s (Nasdaq: GOOG) creds as being the company most likely to violate your privacy.
That said, however, “some of Microsoft’s latest moves have a distinct ‘Big Brother’ feel,” she wrote. “There’s its application for a patent on a system that would allow managers to monitor workers’ physiological states through their computers (ostensibly to offer help to folks when they become frustrated). And then there’s the partnership with MediaCart Holdings that promises to serve up advertisements geared to the position of a shopper’s grocery cart in the store.”
She describes as “creepiest” a Microsoft initiative, in which “sensors gather information about customers as they move through a store. Then the info is analyzed, a customer profile is generated, and targeted advertisements are presented to the customer as he strolls throughout the store.”
How Necessary?
All concludes by wondering how necessary such programs are in the first place. Retail loyalty programs, she wrote, would have more value if companies would do a better job at collecting, analyzing and managing their data. Her son is 7, and All still receives coupons for diapers and baby supplies. “I wish my grocer would tweak its underlying business processes to offer me more relevant ads — but not so relevant that it creeps me out.”
As it happens it will take more than a “tweak” for All’s grocer to be able to offer her relevant adds. Prediction and forecasting technology used by retailers to not only guess what their markets and customers will do but also make the best offer in anticipation of that behavior has been woefully short of the mark.
Technology has not always delivered as promised in this sector, Sheryl Kingstone, a Yankee Group analyst, told CRM Buyer. The good news is that it has improved to such an extent over the last year that the line of business user is able to easily leverage it.
The bad news — from an industry perspective — she said, is that most firms are still using traditional analytics, such as what a customer bought in the past, where and in what quantities, to determine the offer it will make to that customer in the future.” It’s not the most effective way to come to these decisions, she said. And as All pointed out, it’s kind of creepy.
Community Strength
Will Price discusses these problems in his post “Subprime failure and prediction markets,” which starts out by asking the intriguing question: “What do the subprime meltdown and Iraq have in common?” The answer: “Massive failure due to forecast errors.”
The inability for business or government to forecast accurately, however, is not limited to these grandiose failures, he wrote. “Examples … are legion — product ship dates, sales forecasts, demand forecasts, value-at-risk models, elections, stock price predictions, etc.”
We will never be able to predict the future, Price concluded. “However, all of us should consider how we can open up our decision making processes to allow for non-biased, comprehensive input that allows the wisdom of our organizations to weigh in on key decisions - better inputs enable better capital and resource allocation decisions and can help avoid disaster.”
His answer? Leveraging the tacit or explicit wisdom of a community. “Prediction markets are needed to help unlock the tacit knowledge of organizations and to lessen the colossal forecast errors now reverberating through our economy.”
Tom Elrod applied Price’s ruminations to the limitations of CRM tools, in his post, “Forecasting beyond the pipeline.”
“While many of these companies using a CRM are able to produce short term forecasting based off of the immediate sales pipeline, they often have little or no way to forecast beyond it,” he wrote. “The main reason for this is that they usually don’t have the data needed and when they do, no way to analyze that data for predictability.”
Elrod’s answer to the problem was part internal ingenuity and part a call for better tools. He uses as an example a company that has figured out that 90 percent of its sales came from leads that visited its web site at least four times within six weeks of becoming a lead within the CRM application. “Given this indicator of sales, you can start predicting sales from web traffic. For example, if know that have 25 companies who have visited your web site more than 4 times within the last 6 weeks, then forecast a revenue of $562.5K in 3.5 months.
There are a number of data points correlating web traffic patterns to sales patterns that can be used to help forecast future sales, he wrote. The key is to be able to derive these correlations based on historic sales information. “Having a tool that automatically ties web traffic patterns to consequential sales opportunities also helps.
Development Tips
Developer Jane Mantilla offers a few development tips for code customizations for Microsoft CRM 4.0.
“Why am I talking about code customizations and not just customizations?” she wrote. “Well just because in CRM you can do a lot without a single line of code… and this time I want especially to talk about the code customizations.”
Technorati Tags: creepy data, customer service, microsoft crm
