The market for analytics software--already worth $3.5 billion--is expected to experience a 35 percent compound annual growth rate over the next five years, according to Meta Group Vice President David Folger. Propelling the demand for analytics software are major new deployments of ERP, CRM, and other enterprise software packages. Once these systems are in place, managers usually want to get a better understanding of the data that's being entered into them.
Rather than simply facilitate operations and transactions, analytics can help enterprises analyze what's going on. For example, the operational aspects of CRM software might include collecting customer information, updating customer data files, and then routing the customer to the appropriate next step in the interaction, such as to a specific sales agent or information resource.
Analytics go a step further. Analytics software takes that same data, cross-references it, and then analyzes it to help users get a more complete picture of what's happening in the business. Instead of just routing a customer to the proper agent, for example, it might tell the agent that this is a customer who spends a lot of money on a particular type of product. That way, the agent can recommend a new product that the customer is more likely to buy.
With analytics, managers examine key performance metrics in a "dashboard"--a custom view that's often accessed through a browser while data gathering and analysis take place in the background. For example, one of the metrics on a field-service manager's dashboard might be the average number of hours it takes individual service representatives to respond to customer-service requests.
Operational and analytic capabilities are often marketed and sold separately. But that may not last. "The distinction between operational and analytic abilities will probably go away by the end of the year," says Kevin Scott, a senior analyst with AMR Research
Just another name for business intelligence?
In many respects, analytics capabilities aren't really new. They have a lot in common with business intelligence software from vendors such as Business Objects, Cognos, Hyperion, MicroStrategy, and SAS.
Business intelligence can be broken into three functional levels or layers. The first level involves gathering needed data from various applications and databases and turning it into a report that can be distributed to the appropriate decision markers. The second level, known as online analytical processing, cross-tabulates collected data to provide decision makers with a more specific picture of what's going on. The third level, known as data mining, is predictive in nature and is used to forecast and model future performance.
The biggest distinction between business intelligence and analytics lies in the intended users. Business intelligence systems are aimed at analysts, whereas analytics are aimed at managers. Analysts collect data, analyze it, and prepare reports for top-level managers and executives. This process helps companies identify important trends, but it is time-consuming and requires specialists who know how to use the tools and perform the analysis.
The biggest advantages of analytics over business intelligence are the focus and speed they offer to managers. Rather than providing a full range of analysis and reporting tools, the analytics dashboard presents only the information that applies to managers' jobs and that helps them make better-informed operational decisions. Analytics can improve response time by making important data immediately available to managers.
Despite the potential advantages, companies don't always plan and implement analytics capabilities. Here are four precautions worth taking before committing your enterprise to analytics:
Lead with business strategy. Analytics should fit into an overall business strategy. For example, if you want to add analytics to your company's CRM system, ask whether you want to improve customer acquisition, customer retention, or some other aspect of customer service.
Develop corporate analytics standards. Evaluate the application of business intelligence or analytics across the entire organization. Create corporatewide standards rather than implementing analytics department by department--which will only create barriers to sharing information. Use the opportunity to create an infrastructure that spans the enterprise rather than creating more departmental data and application silos.
Map data completely. Analytics depend on gathering and integrating the right data. That data often resides in a variety of systems spread throughout the enterprise, and integrating it can be very expensive. Before giving analytics projects a green light, make sure you understand what data you need, where it's stored, and what kind of integration will be required to create finished dashboards.
Be realistic about deployment times. Leave plenty of time to implement analytics projects. AMR's Scott says that although "you hear about three to six months for an average deployment, the project isn't going to be integrated to anything." He says it takes at least a year before an analytics system is assembled and working.
Although analytics aren't entirely new to enterprise software, they've become more valuable as they've become accessible to managers. Properly planned and implemented, analytics can help managers make better and faster decisions on the spot.
Does analytics software offer companies concrete advantages or is it likely to become shelfware? E-mail Adrian or Talk Back below.
