Business Intelligence

Many data warehouses grow in size disproportionately because technicians do not get to say 'no' to 'excessive' demands of users.
 It is preferred to carry out the project with people from the company, when they do not have time or knowledge to comprehend it.
 Fixing a date of entry into production of the system unrealistic dates and further leads to more delays.
 The budget for the project is small in comparison with the degree of complexity to develop.
 The selection of software and hardware is sometimes done with criteria of general agreements or commitments, rather than purely technical.
 Before the project, not made benchmarks or 'proof of concept to determine feasibility.
 The source data are not clean. Duplication, errors, incorrect characters .. ETL process involving more expensive, larger size of the database and poorer performance.
 The project sponsor does not exercise as such during the year. No 'down to earth. "
 Bad choice of consultants and excessive turnover among them.
 Low end-user involvement leads them to feel some frustracción with the results.
 Fall into the trap of 'computer everything can be done' and start with customization, write code outside of standard features.
 Do not justify the project within a business strategy.

There are many more factors that can crash a Business Intelligence project, but they can do literally 'overthrow', not win more business for consultants, poor product image and internal risks to the CIO and other sponsors.

 OLAP systems. Tips for proper use.

Let us assume that we have undertaken a detailed analysis of business needs has been discussed with all partners and users, we identified the needs of reporing and access, and finally, we clear the model (variables, formulas, dimensions. .) will be included.
It is at this point when we consider the key question: What Store idle method will we use?

We have all the data in our transaction system, which allows faster ride, but may be less efficient. Or we can precalculate information for it to get fast and accurate.

It is a very important decision, because it may involve higher cost of maintenance and licensing.
This is where it should clarify these acronyms:
OLAP online analytical processing.

This is a way of storing information in a database that allows more effective the queries. It is a shortened definition, of course, the reality is more complex.
MOLAP: Multidimensional OLAP. Both the source data as aggregate data or precomputed reside in the same format multidimensional.

Optimize queries, but requires more disk space and different software.

The first point is leaving to be a problem: the disk space is becoming cheaper.
ROLAP: Relational OLAP. Both aggregate and lookup data and source data reside on the same relational database.

If the DataWarehouse is very large or you need quickly by users can be a problem.
HOLAP: Hybrid OLAP: A combination of the two. Precomputed aggregated data and are stored in multidimensional structures and less nievel of detail in the relational.

Requires good analytical work to identify each type of data.

From a practical point of view I would add some other features of an OLAP system:

- Must be fast. Not much time must elapse between the need for information and outcome.

- Must have a functional language and business.

- Must be easy to use, with wizards and templates.

- Must be able to integrate API.

- Must have powerful graphics capabilities.

- Must use maps on a regular basis.

- Ability to store and share the reports and estimates created by users.

- Government should lead the users, not IT.

- The time of implementation (project) should be very short.

- Must generate measurable responses to the decision-making.

- We must be able to obtner ROI with OLAP applications.

As a final summary we can say the three main areas to look are the choice of the people who use the tools of those who take the lead in the project and external consultants.

Besides all this, the system must be within a clear business strategy in the medium and long term, to avoid unnecessary expenses patch solutions.