What is most valuable?
The jobstreams and the ways SQL-scripts can be effectively organized behind icons.
How has it helped my organization?
- It gave the organization a better insight in the quality of the data, which in many ways was poor.
- It gave the call center the possibility to monitor its effectiveness and efficiency a lot better.
What needs improvement?
- The Cognos Data Manager is very datamart/Kimball oriented, and it should be made more flexible, so that it can be used for architectures like Inmon’s or a data vault.
- Some of its components are not very stable (for instance the lookups).
- The prescribed working order (dimensions and hierarchies, then fact-tables) makes implementing design adaptations a complex and risky operation: loosen one thread, and a lot more go with it, and this should me made more flexible.
For how long have I used the solution?
I've used it roughly for more than a year.
What do I think about the stability of the solution?
The stability of the Cognos Data Manager is quite good. It doesn’t crash very often, but of course making daily backups of your work is a wise thing to do, as with any tool.
What do I think about the scalability of the solution?
How are customer service and technical support?
We only had a company provide digital training. Their communication was poor, and they worked with a call center which was totally uneffective. Technical Support
I had to make do with what could be found on websites and the internet. The information IBM gives here is solid, but very technical and hard to read. Other websites have to be handled with care as many contributions are more enthusiastic then useful.
Which solution did I use previously and why did I switch?
I started using Cognos Data Manager because the company I worked for already used it.
To start with SSIS:
- SSIS is a lot more flexible than the Cognos Data Manager. In the latter, the user is pushed into the direction of a Kimball datamart architecture (data marts, which in a later stage are to be integrated into a integrated solution). SSIS leaves a lot more room to the user to make his or her own decisions (and suffer the consequences of these)
- The user interface is a lot more user friendly in SSIS, while Cognos Data Manager is somewhat outdated (but can in capable hands still be used very effectively)
In comparison with large scripts in Oracle and SQL Server:
- The Cognos Data Manager provides a comparison of large scripts for a real and substantial improvement. The icons and jobstreams provide a lot of structure, and are a lot easier to read and interpret than large scripts which can be hundreds of pages long
- Functions like lookups are very nice, but not always enterily stable (which is a serious drawback in their use)
- Combining Oracle and SQL Server databases with each other is a definite strong point of the Cognos Data Manager (although no longer a unique selling point)
- The SQL-scripts are pasted behind the icons. It makes tracing errors and the logic of processes a lot easier
What about the implementation team?
It had already been installed, and I got a lot of support from other people in the company. I have no experience with a vendor team.
What was our ROI?
The company I worked for got a better understanding of its poor data quality, and many inefficiences in its operations. But in their mindset, speed is everything, and precision is often regarded as something of a luxury. They got more insights, made a lot of noise about it, but made no structural efforts for improvement.
What other advice do I have?
- Consider carefully if you want a Kimball architecture with datamarts for your datawarehouse. If so, the Cognos Data Manager is a good option. If not, I think you should reconsider
- Consider what exactly you want to achieve with your datawarehouse. For having a historical database, it can be quite effective. For daily control, in most cases a datawarehouse architecture is unsuitable and unwieldy. Direct reports on source systems are than a better option
- If your organisation has issues with data qualities, a datawarehouse won’t solve these. It can only make these visible (and there are probably a lot of cheaper ways to do that)