It is not news to anyone that information and media has exploded in the last decade, largely as an effect of new technological capabilities and the Internet. Although consistent with past technology trends, most Application Architectures (i.e. how applications are designed and structured) have addressed each information type (structured and unstructured, numbers, text, image, audio and video, as independent concepts) with applications that don’t fundamentally integrate the information types.
There are three basic design approaches that yield the information integration needed for the modern Customer Centric application:
Modern Customer Centric applications are increasingly called upon to apply situationally optimized combinations of these to bridge the myriad types of information in creating successful systems.
For an example, envision an application for a mobile service professional that makes it possible to look into the Customer database to see what product versions a Customer owns (Structured), linked to the schematics of those devices and videos of repair procedures (Media), referencing posts from other service professionals about problems and resolutions for that device, perhaps keyed to specific elements on the schematic (Unstructured).
The power of integrating the information types radically improves the usefulness and usability of that application, potentially improving customer service and lowering costs.
Creating the new application architectures that embrace integrated information types requires new approaches to information design. Traditional data analysis, while well honed for use in Structured data environments, is not fully sufficient because of limitations in describing unstructured information. Fortunately the rapidly emerging practice of Semantic Web analysis and modeling appears to be a methodology that encompasses all the information types, and facilitates discovery of the linkages across the types. When skilled practitioners, with the assistance of the rapidly emerging set of tools available, perform Semantic Web analysis across the functional space and existing information artifacts for that space, it can be seamlessly used within an agile development methodology, yielding benefits to the application design without adding significantly to cost or schedule.
The implementation neutral information design is only the starting point. The technologies that support the information types have developed independently – the new application architecture needs to rationalize and embrace these technologies where appropriate. While application architectures have commonly focused on using of each of these technologies independently, to deliver full advantage of the business benefits enabled by integrating the information types, Architects will need to design to make use of the strengths of each within the context of a shared application architecture, selecting a design approach that is for the business situation (see Sidebar).
There is no single right answer to the question of which design approach is best. Each approach is highly skill set and desired business outcome dependent. Recognizing the balance between the needs for integrated information compared to the potential cost in resources and schedule, in addition to the state of current information assets and the desire to refresh them will lead to the best choice for the given situation. Furthermore, the rapidly changing technology landscape requires building applications that can absorb change in the future.
Despite the potential costs and uncertainty, creating Application Architectures that integrate the information types is the only path to delivering the business benefits of the Customer Centric Applications.
November 11, 2013
Andrew Weiss is a research and consulting fellow of the Return on Intelligence Research Institute. He has served as Head of Technology R&D at Fannie Mae, as Chief Architect and COO of two software firms, and as SVP IT Strategy at Bank of America