An in-campaign modeling process begun this spring by Customerlinx, a customer management service firm in Fredericksburg, VA, has improved response rates for most of its Fortune 1000 clients and increased revenue by more than 30 percent.
“Consistently, we’ve increased response rates for our clients by several orders of magnitude,” said Jeffrey McDermott, Customerlinx’s president/CEO.
The process integrates Decisionhouse customer-behavior modeling software from Quadstone Inc., Boston, with Customerlinx’s inhouse data mart. Customerlinx accesses and analyzes clients’ customer databases, then profiles the customers with the highest propensity to buy a client’s products or services. It does this by analyzing behavior patterns demonstrated during previous campaigns. Customerlinx then tests the model, makes any required modifications and runs additional tests.
“Once we are confident we have the right model, we will go with it full bore,” said McDermott. He added that the system is usually modified, revised and updated daily.
“You are not sending a lot of direct mail pieces to people who have almost zero propensity to buy a product or calling people who have no propensity to buy the product,” he said. “Therefore, you are making more effective use of your marketing dollars. You are going after the people that are most likely to buy the product.”
The technique has helped one of Customerlinx’s telecommunications clients increase its response rate from 0.8 percent to almost 4 percent.
Another reason for the success of the program, McDermott said, is its speed. Decisionhouse interfaces with a variety of data sources, allowing it to load data automatically. As a result, Customerlinx can analyze the data quickly to refine and redirect campaign resources rather than waiting for results from one campaign to improve the next. According to the company, it now takes 25 percent of the time formerly required to complete a company profile.
“Our analytic processes would be four to five times as long without Decisionhouse,” said McDermott. “The faster we can react to the data coming in, the faster we can fine-tune our results. Decisionhouse makes it incredibly easy for our analysts to build a robust and accurate model.”
In addition, Decisionhouse has enabled the company to maximize its staff resources. Without Decisionhouse, it would take four to six additional employees to meet demand.
The modeling process also assists Customerlinx in its new pay-for-performance billing program that bills clients based on how many sales Customerlinx makes for it – either through telemarketing, traditional direct mail or Internet marketing campaign – as opposed to the hours spent on the campaign.
“Determining the truly profitable customers requires an understanding of customer behavior,” said Mark Smith, senior vice president of Quadstone. “Once marketing analysts have this information, they can determine the customers who can be influenced most by a marketing action. This is very important, especially for a company such as Customerlinx that is willing to share in their clients’ risk by basing their entire fee structure on consumers’ response to a campaign. This model gives their customers guaranteed results. They know they will not incur a lot of expenditures at the risk of not getting good enough results.”
