Yesterday QlikTech announced it has completed the acquisition of privately-held integration vendor Expressor. Financial details of the deal were not disclosed.
Is this move a shift in QlikTech’s focus to enterprise ETL? I don’t think so, and in meeting with Donald Farmer, VP of Product Management, and Jeff Boehm, VP of Product Marketing in New York, they were quick to confirm that QlikTech is not pursuing the ETL space. Instead, it’s about meta data re-usability and data governance.
Currently, one of the biggest short comings in QlikView and one in which they lose deals to competitors such as Tableau or TIBCO Spotfire is lack of a meta data layer. The process to extract data from source systems or a data warehouse and load into QlikView’s in-memory engine is based on scripts. Calculations (which QlikView refers to as expressions) beyond the scripts are specific to each sheet within a dashboard application. So if the extraction, transformations, and calculations for a widely used measure such as “Customer Churn” change, that can be a lot of scripts and sheets a designer must touch. Some designers have been able to foster re-usability through shared scripts (shared QVD files), but it’s a work around.
Expressor brings a graphical user interface to the data extraction process. QlikTech is positioning Expressor as a way of doing data modeling in process, iteratively while simultaneously building the ETL and the business application. Keeping in mind that ETL is not my core coverage area, when I visited with Expressor at QlikTech’s conference earlier this year, I was impressed by its ease of use, not sacrificing simplicity for functionality. However, maintaining calculations beyond the initial load is a challenge that QlikTech must still address, one that Tableau took steps to address in its version 7 release last fall. TIBCO Spotfire, meanwhile, has had a meta data layer for years, with its Information Links. Neither competitor has the same degree of scripting and ETL capabilities that QlikView has, giving QlikView a leg up when accessing complex, disparate data, but can be a challenge for less sophisticated users trying to access a single source.
While I think the Expressor acquisition is a good one for QlikTech, the company needs also to address the following challenges:
- Ease of use and maintenance, particularly for the designer. QlikTech has been emphasizing this as a theme in its next version.
- Inability to articulate and position its data marting capabilities, while also allaying IT’s concern about data chaos.
- When to leave data on disk or in a data warehouse versus loading into in-memory; QlikView’s approach is currently all or nothing.
While the Expressor acquisition is a smart-move, it is one that will require QlikTech to address concerns about moves into the broader ETL space and not alienating its larger data integration partner Informatica.
For more detailed reviews and side-by-side comparisons of any of these BI tools, visit BI Scorecard Also, be sure to take the 2012 Successful BI Survey that tracks BI adoption, causes for success and failure, with analysis by BI tools used.