Business Objects recently announced its 'universal data cleanse' initiative, which aims to standardise data quality beyond its narrow traditional niche of customer data to broader data types in the enterprise.
The stated goal is to provide companies with a single, consistent 'panoramic' view of their enterprise data assets, at the same time avoiding duplication and discrepancies of data and data quality tasks. The unstated goal is to evolve Business Objects' information management suite into a full-blown enterprise master data management platform.
Business Objects' universal data cleanse is a direct response to the broadening focus of data quality initiatives. UDC extends the core data customer cleansing capabilities of Business Objects Data Quality XI platform to other types of enterprise data like finance, sales, products, and procurement.
For example, whereas traditional name and address cleansing removes duplications and discrepancies between different name and address formats, capitalisations, abbreviations, and punctuation, UDC does the same thing for product descriptions that refer to the same item but which have been entered multiple times.
UDC is provided as an option for Data Quality Release 2 XI platform, which is based on technology acquired from Firstlogic in 2006.
That acquisition gave Business Objects an immediate foothold in the data quality market with arguably one of the best data cleansing tools on the market, not to mention a huge customer thanks largely to astute licensing agreements with leading IT firms.
UDC is the first major new product release since Firstlogic's data quality software came under Business Objects' wing. It is also the latest addition to an impressive information management stack that also consists of extract, transform, and load, and data federation tools that Business Objects acquired from Acta Technology in 2002 and Medience in 2006. What is missing is master data management, for now at least.
Firstlogic's data quality suite focused mainly on customer name and address data. The initial release of Data Quality XI did more or less the same. But the need for clean and consistent financial and product data is also paramount. Business Objects claims it has been working on expanding Firstlogic's capabilities into these areas for the past year as a direct result of customer demand, so the company claims.
UDC is the first real fruit of that endeavour, and will be a timely addition to the Data Quality XI suite as companies look to expand the scope of their customer-centric quality initiatives to perhaps more problematic areas around financial, product, and procurement information. That means that data quality is no longer restricted by the software to customer data integration.
UDC has to be flexible enough to do several things:
* Parse and reconcile disparate character-string representations of structured data, so that it can accurately identify product descriptions and place data in relevant categories.
* Parse semi-structured data with thousands of characters.
* Process rules associated with weighted logic in data dictionaries.
* Provide tight integration with Business Objects' Insight data profiling tool also gained from Firstlogic.
Internationalisation, traditionally a weakness of many data quality toolsets, also needs to be addressed. UDC does that quite well, which is important for enterprises whose businesses and IT systems span across the globe but still need to interface with one another for knowledge/data exchange. UDS's data dictionaries come preloaded in six different languages.
UDC also addresses the important issue of data quality centralisation, both as a process-driven initiative and as a technology platform.
In the past data quality was usually implemented as a point cure to a departmental or systems-specific data management problem.
The result was silos of use that in turn resulted in silos of clean data. The danger is that one man's clean sheet is another's dirty laundry.
Providing a 'standardised' set of clean and reliable enterprise data is therefore the end-goal for UDC. That means garnering greater enterprise scope for data quality by consolidating these silos of use into an enterprise-wide initiative, which also confers all the benefits of cross-project re-use (cleansing logic, profiling rules, etc) and consistency.
That possibly points to the notion of a centralised data quality competency centre that governs data quality strategy and operational tasks and processes in the organisation. The alternative 'architecture' is to federate data quality across departments, leaving existing systems and processes in place, but at the same time better integrating the data silos.
Customer data quality remains the bread-and-butter of data quality, and will continue to do so for some time. But companies increasingly recognise the need to apply data quality to other areas like finance, product management, supply chain, and human resources that are arguably more complex propositions.
That broadening focus is now forcing traditional name and address cleaning data quality vendors to broaden the scope of their suites, which is what is happening with UDS Data quality XI. It is also interesting that Business Objects is once again adopting the 'universal' tag in its product. A 'universe' is a central concept to Business Objects that refers to a meaningful, business-oriented representation of a relational database.
UDC also sends a signal that Business Objects is keen to establish an affinity with MDM, a technology that has so far eluded the company. With research showing an increased momentum in new MDM projects, UDC, with its broad focus on multiple data types, is a first step toward providing a full-fledged MDM offering.
Business Objects has a collection of data integration, data quality, metadata management, and data federation tools that are suitable for customer data integration and product information management projects. But it still has some work to do to claim full-blown MDM prowess.
Whether UDC gives Business Objects an immediate leg-up on rivals like DataFlux, Informatica, and IBM Ascential is debatable. Many of these companies have already embarked on product strategies that are evolving their data quality suites beyond staple customer name and address cleansing, which is where the bulk of the money is to be found in data quality.
But with data management initiatives like data warehousing and master data management, and compliance now starting to take a more ambitious 'enterprise' view, data quality suites will have to spread their wings to other types of enterprise data.