Grasping mashups in BI

The dissemination of traditional information is about as interesting as a dry piece of toast. For the past decade, business intelligence (BI) software companies have tried to spruce things up a bit with fancy graphics and dashboards, but have succeeded with only marginal triumph.

However, emanating from the Web comes along a new information visualisation technique that is changing the way information is presented. This is called mashups, and mashups that are specific to the BI space are referred to as bashups.

The term mashup refers to an application that is built by combining two or more data sources, typically using a graphical representation, and is one of those radiant concepts that only comes around once in a while. Bashups within BI typically refer to the mixing of BI information with non-typical visualisation (enriched visualisation, if you like), allowing cross-functional analysis to take place that will allow for a better understanding of relevant information that impacts business decisions.

An example of this would be to map product sales per region on top of weather abnormalities on a Geographic Information System (GIS) map, to ascertain correlations that may not otherwise be apparent. Bashups become a powerful technique when enterprises start presenting this data beyond grids and charts. Such an example of the combination of GIS and BI has been said will become important in the future, and as such, a general understanding around bashups is crucial going forward.

Let's take a look at common mashups - combining Web 2.0 with BI 2.0. BI 2.0 is a term that encapsulates several important new concepts about the way information is used and exploited in businesses, organisations and government. The term is also intrinsically linked with real-time and event-driven BI, but focuses on the application of these technologies on business processes.

Over the past few years, companies have started to present their data warehouses as Web services for use by other applications and processes connected by service-oriented architecture (SOA), or middleware such as an enterprise service bus (ESB). Many mashups that are available today are hosted as sites on the Internet or Intranet, providing visual representations of publicly available data - thus mashing up BI and the Web.

Earlier versions of mashups were known for being tedious and time-consuming. However, custom scripts on the server side of things are starting to be replaced by standardised libraries that will automatically generate the required client-side script. We are also seeing standardisation in the message formats. The creation of mashups was once only the domain of the developer, but there is now a movement that has been put into place to set the ability to create mashups directly into the hands of the end customer. This has resulted in an increase in the importance of common schema and metadata in the development of mashups. As the ability to create a mashup becomes more viable, implementing systems and procedures that will allow for a variety of mashups to be created using such systems, presents an opportunity that industries should definitely take note of.

If we take a look at the call centre industry, for example, BI has already been implemented in most call centres today, and the reasons why are obvious. However, in addition to just looking up the customer information, developing a mashup here will allow for this industry to plot a specific phone number on a map using a publicly available service, and will also be able to display a list of local service centres or parts suppliers for products overlaid on the map. With this data in hand, one working in a call centre may be able to answer the customer's questions in seconds - imagine the customer satisfaction and the relationships that can be built!

Gartner has predicted that by 2012, one-third of analytic applications applied to business processes will be delivered through coarse-grained application mashups.

Gerhard Botha is principal consultant at PBT.

Two examples of mapping services that can be used in developing a mashup are Google Earth and Microsoft's Virtual Earth, both of which have map views and satellite photographic images of the Earth's surface. If your data includes spatial data as latitude and longitude (the locations of your stores or customers) then these can be sent to the API and surfaced as points on a map. This will enable one's business to combine business data with one or more of the externally available services, giving a workforce, or even customers, very detailed maps displaying distribution centres, store or rental locations, sales in different territories, etc.

Gartner has predicted that by 2012, one-third of analytic applications applied to business processes will be delivered through coarse-grained application mashups. There are a multitude of mashup tools available that allow for the creation of mashups. For example, Attivio's Active Intelligence Engine (AIE) indexes both unstructured content, as well as transactional data and preserves the normalised form of the transactional data in the index. Their technology allows one to join across text and data so that the mashup of numbers and text can take place - allowing for the discovery of patterns, correlations, and important new insights about an organisation's performance using whatever presentation layer is required to connect to AIE.

Another great mashup tool is that known as Dapper, which is an online mashup tool that takes the concept to the extreme, making it possible to convert and re-use just about any source of information on the Web. Along with these tools may come others, all of which represent great mashups of various components helpful in business.

Although a somewhat different idea and concept, mashups can provide businesses with the opportunity to determine information essential not only to conducting efficient business, but for crucial business decision-making. Developing mashups within the BI space, ie, bashups, can push the BI concept to one that is useful, and as such can promote utilisation by more individuals of BI within an organisation.

* Gerhard Botha is principal consultant at PBT.