Pitfall # 1- Ignore Business Intelligence requirements
and use up 90% of project resources for technical implementation, data modelling and transformation logic. Leaving only 10% for analysing the required Business Intelligence application is simply not enough. You will end up with a technically state-of- the-art Data Warehouse of no use for your special Business needs

Pitfall #2 – Extensive and longterm Analysis,
specifically requirement analysis are important, but require a lot of time. In practice you never analyse all requirements before starting design and implementation. Waiting for all possible requirements to be analysed in detail will postpone design and implementation phases and even further delay any real results from a Data Warehouse implementation.

Pitfall #3 – Many tools
Data Warehouse development often make use of a lot of tools. This results in the problem, that code and documentation are spread over various systems and formats. Transporting information over different formats and media is error prone and will lead to errors.

Pitfall #4 – Offer everything, not what is required.
A Data Warehouse that collects all data, is not necessarily a useful system. Having all Data in the Warehouse does not mean that all Data are required or can be used for anything at all. There is a high risk of amassing useless data. A well-thought Data Warehouse is a basis for performant Business Intelligence applications. This will keep the Data Warehouse useful and interesting in the long term.

Pitfall #5 – Long development time
The longer it takes to analyse reuirements, design, model and create transformation rules, the less willing project participants will get to finally use this Data Warehouse. Avoid a project stop for cost and time reasons – meaning frustration reasons.

Pitfall #6 – Disregard changes to come
Supposing end users really know what they will want in the future is risky. During project design end users can decide what kind of reports and reporting funtionality they need. Six month later the same users will want more. Keep your Data Warehouse architecture open for change and further development.

Pitfall #7 – Get too technical
Finding the optimal technology is not the first priority of a Data Warehouse project. Most important is the longterm acceptance of a Data Warehouse by end users and Business Intelligence developers.