Innovative product suites convert the promise of SOA – making systems interoperable – into business processes that span enterprise boundaries. Mash-ups, Websites or Web applications that recombine content from different sources to create enterprisewide business scenarios and comprehensive forecasts, allow a new integrated experience. With the help of the Web services, these applications mash up weather or traffic data with information from SAP NetWeaver Business Intelligence, mySAP Customer Relationship Management (mySAP CRM), and third-party service providers, such as market research companies. The information is then superimposed on satellite imagery. The resulting composite, or mash-up, can provide compelling 3-D insight into business ecosystems.
Visualization close to reality
The use of SOA has been particularly helpful with geographical information systems (GIS). From finding a retail outlet to tracking a package shipment, the use of GIS Web services gives companies a realistic picture of the business environment. Google Earth delivers a platform for displaying and analyzing spatial and geographic data. Google Earth works together with numerous GIS. For example, Web services from Environmental Systems Research Institute (ESRI) provide information on latitude and longitude. Yahoo offers Web services that deliver cartographic data on traffic flow. When the information from both providers is combined, Google Earth can display the traffic flow. This simple principle can be used to display complex business processes geographically. The result is SOA related to geographical data that companies can use to develop a transparent CRM strategy. The following three steps show how they can achieve it.
Step 1: Capturing information on customers
Information on customers is indispensable to successful CRM. For example, the GIS from ESRI – as part of the mash-up – helps by visualizing the spread of customers throughout a geographic region. If a retail company wants to open a new retail store in a new city, the ESRI Web service provides information on the typical spend pattern, demographics, and the size of the commercial area. Evaluations of this information can help determine how profitable the new retail outlet would be.
BPEL (Business Process Execution Language) supports the selection of data from various SOA service providers like Yahoo, Google, Amazon, or SAP databases. The ecosystem visualization solution, Magma Ecosystem, from Enterprise Horizons, for example, uses BPEL to unify a seamless dataflow on a geographic platform and to present a seamless stream of information content. The Magma server that handles the external requests and returns the results in binary format is programmed in Java and runs on several hardware platforms. As part of the mash-up, it first makes a Web service call to the ESRI demographics engine.
The demographics engine enables companies to access and manage enormous spatial data sets, such as customer locations or property parcels. Supported by the demographics engine, companies can receive information on typical consumer behavior by region or postal code. The Google Earth client tracks the information visually, so that the distribution of customers in a region can be imaged. For example, people around a university tend to reflect a consumer demographic of young people, whose spending patterns differ greatly from those of baby boomers. BPEL can be used to call Web services on the Magma server to obtain data like annual income, wages, and shopping patterns. The Web services provide the demographic data that the server uses to visualize the data in Google Earth.
As another part of the mash-up solution, the Keyhole Markup Language (KML) server from Google Earth renders this information and overlays it on the satellite images of Google Earth. KML is an XML-based file format that was initially designed to allow Google Earth users to exchange map annotations easily. The initial purpose has now been extended to allow users to translate traditional GIS data into the KML format.
Step 2: Understanding the market
Customers and consumers behave differently. Clustering customers into target groups focuses marketing and sales efforts and often results in greater productivity for a specific channel. One important parameter that helps cluster customers is geography. To analyze customer behavior by different regions and then personalize a sales strategy to suit target consumer groups, an enterprise must have the appropriate data. But performing powerful analytics can be challenging because the required information, such as the history of customer purchases, competitor prices, or the location of retail outlets can be spread across diverse data sources. By using BPEL and a BPEL engine, disparate data sources like the Yahoo! product rating, Amazon product pricing, and ESRI’s demographics utilization can be combined to create new processes.
For example, ESRI has identified an exhaustive list of customer segments, called Community Tapestry, for a typical consumer products company. Segmenting consumers based on geography enables retailers and manufacturers of consumer products to launch targeted promotion plans, campaigns, trade and promotions, and other sales incentives. In addition, data from back-end systems and business intelligence solutions like SAP NetWeaver Business Intelligence provide information on the competition, the market, and distribution channel profiles.
Step 3: Simulating consumer product interaction
Products are introduced to consumers in several ways: newspaper commercials, television commercials, billboards, or other outdoor advertising. Enterprises obtain the locations of billboards in various postal codes when they use outdoor commercial service providers. Using the mash-up framework, companies can also obtain information on the average traffic speed on a highway in certain postal codes, on typical billboard viewings, or on total consumer exposure time to the billboards and products. The SOA solution then reads the consumer demographics information by region and analyzes the costs and benefits of investing in specific commercial campaigns – like placing a billboard in view of a sporting event.
Enterprises must evaluate optimum product placement to maximize merchandizing effectiveness. The BPEL solution uses the SOA ecosystem to visualize microgeographies with the same effectiveness as regional analytics. Companies can therefore visualize retail layout and consumer traffic by aisles. When enterprises mash up data from POS systems and data from back-end enterprise resource planning (ERP) systems, they can visualize consumer traffic data in a store. POS systems can also provide information on what people buy at a supermarket. The information can include data from credit cards, shopping cards, radio frequency IDs (RFIDs), and other sensor data sources that the retail outlet can provide. Aisle sensor data can also be used to generate traffic data, which can then be used to visualize shelf visibility to end-consumers in a retail outlet. Some supermarkets are piloting more sophisticated technology such as video cameras that profile consumer traffic for more insightful data on traffic.
Thanks to SOA, every company can capture demographic or statistical population data, the layout of power lines, and the location of shopping centers. Data on traffic patterns in a suburb can provide information on consumer density. With this information, companies can simulate and study the purchasing power of the residents. And 3-D models offer important information on customers, products, and locations even before a company implements a sales strategy. That’s the time for the company to ask if building a new shopping center is worth the effort. That’s now mash-ups and SOA can increase a company’s profits.