On 25th January open source vendor Talend announced the general availability of Talend MDM, an open source master data management (MDM) product based on their earlier acquisition of French MDM technology Xtentis from Amalto. As I commented when that acquisition occurred, this is a significant step, since customers who are thinking about dipping their toe in the water with MDM can now experiment with a free "community" license of some modern MDM software. If customers wish to move beyond this version, then the enterprise version has additional functionality (e.g. around workflow, multiple roles, and of course technical support), and is likely to be priced at under $100k for an annual license. Given that the software of major players in the MDM market typically sells at half a million to a million dollars, this is relatively affordable.
Functionally, the enterprise version of Talend MDM has the features that you would expect from an MDM product; it is based heavily on Java and Eclipse, and uses JBoss for event management. It has built-in data quality functionality (profiling, name and address matching) as well as workflow capability, role-based security and the MDM hub itself. It is able to take advantage of Talend's established open source integration technology for data movement e.g. for bringing together master data sources, and publishing of "golden copy" data to other systems. One important aspect of the technology is that it is heavily model-driven, and does not have a fixed data model, so is inherently multi-domain by design. It is anticipated that the open source community may develop specific "starter kit" data models and integration jobs, but our research consistently shows that customers require hubs that can deal not just with customer and product data, but with other types of master data such as location, asset and financial hierarchies, as well as industry-specific data such as oil-well information, or diseases in the case of the life-science industry. We believe that this "multi-domain" approach (rather than building around a fixed data model) is the one that most customers desire, but frequently cannot get, so Talend MDM is well-positioned here.
Talend will need to demonstrate that the unusual XML database platform that is used (which has some advantages in data modelling and search) also scales properly when significant volumes of data are to be managed, and will need to handle objections from database purists who prefer to standardise on a major relational platform. They also will need to back up their claims of rapid implementation with properly documented case studies. Amalto did have a dozen or so customers, mostly in France, but Talend has significantly enhanced and enriched the acquired technology and new prospects will want to see growing evidence of wider deployment of the new and improved open-source platform.
Overall, it can only be a good thing for the industry to have an open-source MDM choice available; this provides useful pricing pressure on the current industry leaders, and allows customers to get started with MDM in a low-risk fashion.