1to1 Magazine - November/December 2008 - (Page 23) “Obviously a few don’t get delivered on time and some are lost one-to-two-times-a-year model,” he says. “You’re left wondering if completely; that’s just a fact of life. those conclusions are really reflective of what’s going on.” “Janet’s team found that those customers who have problems are And for her part, LeBlanc says, “It’s still a bit early to determine less delighted that those who don’t—which is obvious,” he all the results.” continues. “But she further found that if we are able to resolve his case completely within five business days, that customer would then Expanding the project scope give us a high score on problem resolution, and emerged as three CP execs initially took on the linkage analysis project “pretty times as happy as customers who had no problem. If we messed up much on faith,” LeBlanc says, but now strongly believe in its value. but fixed it, they had more confidence in our service than someone Lest anyone think there’s a given formula for proving—or who’s never had a problem.” disproving—similar links in their business, however, she says some For VentureOne customers, it was proven that their attitudes factors need to be kept in mind. were affected in response to organizational actions like pricing “Some companies can use the same customers year after notification, campaigns, problems, and so on, LeBlanc says. “It was year to measure customer value, and to monitor year-to-year also observed that this small-business customer segment was increases,” she says. “That’s very easy to do in telecommunications, directly impacted by our actions almost immediately, with no lag for instance: They sell you a cell phone, you sign a contract effect; conversely, with larger customers for Year X, get a monthly statement of their behavioral changes are realized charges, then you leave them and go to over a longer period of time. We were someone else. also able to link loyalty to revenue, “But direct mail companies may do a big “We went in with the idea identifying significant potential oppordirect mail campaign in March 2007 and that the more data tunities for Canada Post by continuing [then] in July 2008,” she continues. “It’s not a we have, the to improve the customer experience.” regularly reoccurring pattern.” stronger the LeBlanc says that CP’s parcels Nevertheless, she says, CP has completed a business has increased its percentage linkage analysis on its ad mail segment. “We results will be” of loyal customers by 6 percent saw similar results and linkage to revenue as since 2005. She declines to give we did with parcels,” LeBlanc says. Janet LeBlanc, out revenue numbers. As for regular mail, CP remains Director, Customer Value Management, a monopoly—but that’s not to Canada Post Not all smooth sailing say that the company doesn’t care LeBlanc identifies four primary challenges about its customers, she says. An that CP faced in putting together the analysis of regular mail linkage analysis: 1) the restrictions that customers is currently being legacy systems placed on the collection completed. and manipulation of data; 2) the fact that “We’ve always been far CP customers often switch between more focused on our competiproducts, which complicates accurate tion and our customers than a lot identification of which customers have of other companies [are],” O’Brien actually grown, eroded, or been lost to the says. “We’ve always had a public competition; 3) complex relationships that service mentality.” are difficult to identify within the data, including mergers, He too feels that the linkage analysis of bankruptcies, and acquisitions; and 4) developing results that are the parcels business will be effectively statistically different when the population of customers in repeated with ad mail and regular mail. a particular segment can be small. “This is all good for our customers,” he says. Market Probe’s Pflughoeft underscores the importance of that last “In parcels we found that we were meeting variable. “If you look at this kind of data from an aggregate level, the customers’ basic needs; now we’re looking at relationship might not exist, but if you break it down into quarterly where the subtleties are and what needs reports the relationships become clearer.” Time is also important, he fine tuning.” says: “You need to have enough data so the result can’t be explained away by coincidence. It’s generally nice to have three if not four years’ ONLINE EXTRA: worth of data for what we were looking at in this instance.” To read about how Canada Post links customer Pflughoeft notes that the CP linkage analysis began with tracking activity to business results, go to data “once or twice a year” before going to a monthly model Visit: www.1to1media.com/links/canadapostresults.html toward the end of 2006. “It’s harder to use those data points in the November/December 2008 23 http://www.1to1media.com/links/canadapostresults.html
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.