Question of the Week

3 important development principles that im going to go over, from the point of view of the designer of a dashboard, not the Targit application itself. Thus how does the Targit application allow a designer to adhere to these principles.

- KISS (Keep It Simple, Stupid):

• The criteria bar could be simplified for end users by allowing designers to add custom slicers (not time dimension). A menu with bookmarks is a start, but there is room for improvement.
• Natural language processing to ask questions about the data is another step forward, although it can also be a pitfall.
• Some calculations need to be done in the data source because they become too complex in Targit.
• The set of charts is complete, though no overly complex visualisations, stupid 3d effects, etc.
• the data source NEEDS to be simple and perfect in order to work in Targit. Complex data sources that are not proper star schemes are difficult to work with in Targit. Also any data manipulation needs to be done in the source. Whether this is a good thing is up for discussion

- DRY (Dont Repeat Yourself):

• Calculations that span multiple visuals need to be repeated.
• color consistency, same members get the same color, no matter the measure, which means you could put a single legend on the entire dashboard (with a cleverly formatted bar chart)
• reusing menus
• templates
• layouts get adjusted simultaneously

- YAGNI (You Aint Gonna Need It):

• Targit is sufficiently feature rich to make from the most basic dashboards to a whole intricate network of analysis pages. Therefore this is entirely up to the designer. Targit does not guide the designer in good BI design principles and allows for very complex, cluttered and/or ugly dashboards. In that sense, i feel the Targit eLearning could be extended further than simply learning how to use the tool.