Drilling into the detail: CRM (part 3)


17 Mar 2003

Amid the doom and gloom of the CRM software sector, one area continues to grow and grow fast – analytical CRM. Research firm IDC last year reported that European companies increased their spending on analytical CRM by 30pc in 2001 and expected investment in this type of software to almost double between 2000 and 2005.

The 800-pound gorilla in the field is the world’s largest privately-held software company, North Carolina’s SAS Institute, which, IDC figures show, has a 26.7pc share of the analytical CRM market in western Europe and also dominates the global market. Other important players include Chicago-based SPSS and Ireland’s own Norkom Technologies.

Analytical CRM is the granddaddy of the CRM family, except that when it emerged in the late Seventies it wasn’t known as CRM, but as data mining or data warehousing. In its purest form, analytical CRM consists of high-powered software algorithms that can number crunch millions of pieces of information and out of the morass of data distil trends and usage patterns. The software is used mainly by service organisations such as banks, telecoms and retailers, whose customer bases run into the tens or hundreds of thousands. In this sense, it is a niche application although a notable trend of recent years is for large CRM sellers such as Siebel and SAP to bolt on analytical modules to their existing products and offer it to their existing client base.

The use of analytical CRM varies and partly depends on the nature of the company in question. For example mobile phones companies, for which customer churn – ie, subscribers defecting to another network – is a serious problem, use the software to predict which customers are likely to switch to a competitor so that the network can take measures to retain them. This was the objective of MTN, the largest mobile operator in South Africa, when it implemented SAS Institute’s Enterprise Miner CRM product. “They ended up cutting their prepaid churn rate by 45pc because they were able to identify those customers that were just about to leave and target them,” says Patrick Durkin, country manager of SAS Institute Ireland.

Banks do not suffer churn to the same extent; their problem is more that they have large numbers of dormant or low-activity accounts. For them, analytical CRM provides the means to identify the least valuable customers in order to target them with specific offers in order to drive up the revenue yield from each. It can also identify the most profitable customers so that the bank can cross-sell insurance or a mortgage to someone who had previously taken out a car loan and made all their repayments on time.

Analytical CRM is generally seen as having two core elements. One is data warehousing, which gathers all relevant customer data into one ‘information silo’ giving a single view of an organisation’s customer base. The other is data mining, which yields specific information about customer behaviour – what do they buy, how often and so on. “For organisations trying to service the needs of customers, they need to understand not what the customer said they would buy but what they actually did buy,” notes Durkin.

To these two, he would add a third: a campaign management tool that using the results produced by the first two steps, allows businesses to execute promotional and marketing campaigns. In the case of SAS’s CRM solution, these step components are integrated within a single product suite. “Any of the three can be separated out if a customer requires it but the real value emerges when all three components are linked up end to end,” believes Durkin.

Although a powerful and effective tool, analytical CRM needs to be properly supported within an organisation to be effective. According to Frank Cole (pictured), director of CRM firm QMS, the software needs to be coupled with human analytical skills to deliver maximum value. “With analytical CRM, you need to employ people – statisticians – to interpret data and advise management. My view is that you’d really have to understand the data or else you won’t get a proper sense of what the data means.”

In addition to providers of high-powered number crunching software, which traditionally defines analytical CRM, vendors such as Crystal Reports, Cognos and Business Objects have muscled into the area with powerful data mining software that can feed into business decisions. These products provide regular reports in the form of graphical ‘dashboards’ which, for example, allow a service engineer to view the recent service history of the customer they are about to visit or a salesperson to forecast the sales pipeline coming from the company’s 10 largest customers.

The growing sophistication of these relatively low-cost packages means that CRM software with moderate analytical capabilities is now coming within reach of medium-sized companies. For the foreseeable future, however, high-end analytical CRM applications are likely to remain the preserve of organisations with large customer bases.

By Brian Skelly