QlikTech's QlikView reporting and analysis tool is among a new class of Business Intelligence (BI) software tools. As Ben Harden reported in a recent blog post, BI vendors like SAP, Microsoft, and IBM have traditionally sold "to the IT enterprise, but companies like QlikTech and Tableau are targeting the business and bypassing IT. Their tools are quicker to stand up, more intuitive and don't need the configuration, support, and hardware that the bigger players require."
A Quick Overview
At first look QlikView is fairly accessible to those experienced with BI tools. A ".qvw" QlikView file contains three classes of user-facing components: a script-based data integration language that runs when the user requests a "reload", a data modeling component that looks deceptively like a relational data modeling tool, and a familiar array of data visualizations: graphics, charts, lists, etc.
The scripting piece features the usual native and generic data connections, so QlikView can pull data from the usual relational databases and file sources. The data access component wraps SQL commands within a reasonably robust scripting language that enables renaming, reformatting, grouping, control flow, and calculations to produce "tables" that QlikView uses in its display components.
The relationships among QlikView tables is "associative" rather than relational, so lines between tables don't reflect foreign key relations. Instead, the lines mean that there is a like-named column in each of the related tables. Therefore, data modeling and optimization in QlikView takes some getting used to for the experienced data modeler. Another optimization point: when the user requests a reload the scripts load tables into memory, in effect creating an in-memory datamart that the charts and graphs access. So QlikView benefits from addition of more memory and, assuming sufficient RAM, is very fast after the initial data load.
Visualization elements like charts and graphs draw data from the tables, and are integrated together so that a selection in any one visual element limits the data displayed in all the others. So if the user selects Minnesota in a table of sales totals by state, every chart and graph on every tab in the open .qvw file changes to display data for Minnesota.
QlikTech provides ample resources for getting started, and with a critical mass of installations across Europe, supports a vibrant and knowledgeable user community. Here's one way to get started:
- Download the free Personal Edition: the fully functional Windows-based Personal Edition of the desktop product enables all user-facing features but won't open files created by licensed users, and vice versa.
- Go through free online intro training: There are a number of free training classes available. Start with the Designer and Developer introductions. These two classes consist of about ten 20-minute units each, supplemented with exercises using a .qvw download that provide solid understanding of the structure, mechanics, and some of the potential pitfalls of the product. After the intro there are additional free courses on system management, server-based deployment options, and more. Free sign-up at the training site is required.
- Explore the community and "technical briefs": QlikTech presents a wide array of technical papers on administration, management, and other topics, and a vibrant community site. The "New to QlikView" page on their site serves as a useful portal with links to different entry points, a log of recent user discussion content, and links to content from top discussion contributors.
Of course there's much more to the product and its use, configuration, and deployment than a quick intro will provide, but QlikTech's startup resources provide a useful intro for IT professionals and business people alike seeking quick in-depth exposure to the product, potential pitfalls, and secrets to success.