All Business Intelligence solutions attempt to provide answers to business questions.
However, finding an answer to business questions is often complex and multi-faceted. Most BI platforms can help a user find the answer to the first question they come up with, and possibly a limited follow-on question through simple, single-value drilldown or hard-wired parameter passing. Most BI technologies allow you to view results in a standard report format or perhaps in some very attractive visualizations?though this tends to be where it stops. Ultimately, end users cannot use traditional BI solutions to support a natural thought process with a non-predetermined path of questions and answers.
QlikView is different from traditional BI solutions in that all analysis in a QlikView application is automatically associated. This is based on the data model held in memory which inherits many important relationships from underlying databases and also allows the creation of new relationships to other, previously unrelated data sources. This capability means that QlikView just works as your mind would expect it to.
A question I always ask myself when reading about other BI technologies is whether they really support the analysis of relationships in data derived from more than one data source, not simply the merging together of data at a query level and then display of the results in a report.
Let me expand on this a little. Imagine I have a chart that shows me product sales, sorted by sales value. Within QlikView I can drill into or select products within the chart?though it doesn't stop there. Suppose I also have another chart that show sales trends over time. Now, if I select my top five products I then see the sales trends for those top five products. Conversely, I could start by selecting a time period in the chart of unusual activity, and instantly see the products that have contributed to this in order of importance.
I'm sure the mathematicians amongst you could come up with some sort of formula which would approximate how many different routes of questioning an end user could take through a QlikView application given the number of different analyses available. However, for now, imagine you have four pieces of analysis as shown in this video.
In QlikView, all the relevant pieces of analysis are associated. Because all of the objects in QlikView are completely interactive, supporting single or multiple selections and/or drilldown or drill across, end users have a completely ad hoc ability to ask the questions they want, and also as many follow on questions as required.
This video shows a fairly simple example but even in this case, you will hopefully begin to realize that the QlikView associative experience provides a big step up from traditional query/report paradigms that provide simple prompts with no particular context, and can generate an answer to a question but provide very limited scope for subsequent analysis.
Through the use of a completely interactive interface where all analysis is associated, driven by the underlying data model held in memory, QlikView opens up a new world of possibilities to end users. They can see the bigger picture, ask whatever questions come to mind, and ask follow-on questions by interacting with data in multiple charts or tables?each having its own important window on the data. Users can make selections, drill up, down, or across, sort data in various ways, and change chart types. They can even create their own new views and visualizations which will contain data that is automatically associated.
This unique capability, in conjunction with QlikView's comprehensive search, advanced visualization, and speed-of-thought analysis, makes it clear why end users who have been frustrated by shortcomings in other BI products become such big fans of QlikView.
We would of course welcome your thoughts on this and especially examples/experiences where this aspect of the product has demonstrated value to you.