SAS helps HP understand customer behaviour with automated data capturing, cleansing

Hewlett-Packard (HP), one of the premier IT infrastructure companies in the world, with $72 billion in revenues, 140 000 employees and a vaunted history of innovation in computing products and services, has turned to SAS to automate its process of data capturing and cleansing.

To maintain its leading position, HP's Enterprise Systems Group (ESG) North America, needed to identify market trends in the purchase of HP products and services, exploit those trends and develop better, more effective go-to-market strategies.

"The company needed to break free of manual data entry, manipulation and reporting. It turned to SAS to assist it in automating the process of capturing and cleansing the data for faster, more accurate results," says Annemarie Cronje, solution architect, SAS Institute in SA.

She was commenting on a recent case study released by SAS globally in which HP stated that a four-month trial of the SAS data cleansing and data warehouse solution worked so well that the company is now getting ready to roll it out on a permanent basis.

"Strong market information is critical to a company's planning process. It guides decisions on where to spend marketing funds, suggests which markets to pursue and determines which ones to leave. It helps the company decide how well its sales staff is doing and how to tailor its marketing messages," says Cronje.

The case study quotes Craig Pampeyan, who is responsible for strategic planning for presales at HP. "SAS really helps us understand our customers and see where we can successfully cross-sell and serve our customer needs."

Because data is sourced from several outside platforms, HP needed to reconcile that information with its own more detailed internal material on who is buying what products. The process was complicated because no two sources gather data in quite the same way and customer ID codes are not always uniform.

For example, one customer might be listed under several different customer numbers based on multiple address locations.

"One month we would do manufacturing industry revenue. Another month we would do telecom revenue," Pampeyan recalls. "When we got around to adding up all the industries, the pie ended up bigger than 100%."

Pampeyan needed a complete solution that would encompass the entire scope of the project. So he turned to SAS.

"Other companies couldn't give us an end-to-end solution and we didn't have time to manage multiple vendors. SAS went out of their way to make sure we were covered on every aspect of the project," Pampeyan says.

The key to the SAS solution is its ETLQ (ETL to the power of Q, quality) process that integrates data quality and cleansing into the data loading and warehousing. SAS automates the flow of information from disparate systems, integrates it and cleans it, all in one step, resulting in data that is accurate and easy to use.

Cronje says the results were outstanding. Once the solution was installed, HP could run a complete report - on every industry - in a couple of hours as opposed to the two weeks it took to do one industry in the days before SAS.

"More than creating a holistic view of its market, the SAS solution gives HP the kind of detailed reporting required to see a complete picture of their customers, with eye-opening findings.

"HP now has the ability to see how customers buy their entire portfolio of products. For example it has discovered a correlation between certain printer purchases and certain server purchases, a connection that would not have been easily uncovered from the old way of reporting," she says.