Compared with many other tools it's quite simple to start with Qlik especially if the datamodels are simple with only one/two tables. But this advance even increased if the requirements became more complex.
It's already possible with only little know-how to develop quite sophisticated data-models. Basis is the associative data-model from Qlik whereby all tables will be connected through key-fields whereby synthetic keys and circular references should be avoided. Thoughts about star- vs. snowflake-sheme, normalization and so on can be widely employed back - it's easier as it looks like:
Circular References
The Table Viewer
Synthetic Keys
Merging Data from Multiple Sources
Fan traps and Chasm traps
To Join or not to Join
Don't join - use Applymap instead
Concatenate vs Link Table
Fact Table with Mixed Granularity
If data don't be loaded from databases else from flat-files they have often problematic data-structures including the lack of data. Beside the following links is the inbuilt file-wizard from QlikView a powerful but not well documented tool - helpful is here the book: QlikView 11 for Developers which is in general very much recommended:
The Crosstable Load
Counters in the Load
Generating Missing Data In QlikView
Most often worked the loading from fields without any problems - if not are these links helpful:
QlikView Quoteology
Automatic Number Interpretation
On Format Codes for Numbers and Dates
Further very helpful is the use of a master-calendar within the data-model:
How to use - Master-Calendar and Date-Values
Here you will get further important topics: Advanced topics for creating a qlik datamodel andMore advanced topics of qlik datamodels.
Beside the above used links you will find many interessting postings here within the qlik community to these topic - the notes here are a good starting point to go further.
Have fun!
Marcus Sommer
ps: within the attachment is a german translation - deutsche Fassung.
