QlikView was born in 1994 in Lund, Sweden. While it was quickly spreading in Europe, it was virtually unknown in the US for many years. In 2000, there were few companies that used this BI tool. (I was lucky to work in one of these companies and learn QlikView in early 2000.) The situation is quite different now. Certainly the product has changed a lot, but the most important features that are responsible for its success remain the same.
First of all, the exceptional ease of use. User doesn’t need to learn anything “BI-specific” or any particular technology. If you know how to click – you can use QlikView applications. Well, maybe 15-minute presentation is needed to explain what this particular application is about and how to use it.
Second, there is no steep learning curve for the developers. After initial training (two or three days), you can create QlikView applications. Not the advanced ones yet, but quite reasonable and useful. The only technical knowledge I consider a pre-requisite is SQL, and you don’t need to be on an advanced level, basic “SELECT … FROM …” is good enough to start from. Even if you don’t know any SQL, you still will be able to develop QlikView applications using wizards that will create simple SQL queries for you. Developer certainly must understand the business needs, and know where to get the data which support business requirements.
Another important advantage of QlikView – it is “in-memory analytics” tool, I believe the first of this kind. All data is in memory rather than on the disk or other storage, and this allows get the results fast. Certainly it was somewhat a limitation in early years because of the limits of 32-bit architecture and memory price. Now with much cheaper memory and 64-bit systems, the “in-memory” applications can handle larger amounts of data than ever before.
Unlike the most (if not all) other BI tools, QlikView does not require data warehouse. Data can be read directly from the original sources, and all ETL work can be performed within QlikView application itself in the data load script and on the front-end level. If a data warehouse exists already, QlikView can use it as well.
Speaking of data sources, QlikView can load data from the variety of the RDBMS (I used it with Sybase SQL Anywhere, Oracle, MS SQL Server), Excel files and many file types including CSV and XML. There is also a proprietary data type, QVD (for QlikView Data) for temporary storage and retrieval, which is extremely fast. In practice, it is not uncommon for one QlikView application to read data from several heterogeneous data sources – multiple databases, maybe even of different types, from QVDs and flat files.
QlikView continues changing, adding more features, improving performance, opening new possibilities. The important event in the company’s history were moving to the US in 2006, and going public in 2010. I guess it is here to stay and increase its market share.
Disclosure: The company I work for is a Microsoft Partner