Promoting the warehouse

[mselway]

Hi,
I have been building our warehouse (for an insurance company) for 18months.
We now have a functioning system built along the Kimball methods on MS SQL platform.
But we still have a stack of users on the old reporting systems (a copy of the OLTP system).
My challenge is to promote the new warehouse and improve uptake.
The Bus Matrix is a good starting point to get peoples interest, and diagrams of the datamart are good too.
Any other snippets of wisdom?

[elmorejr]

I've been here before....

The users will need to see what the benefits are for switching over to the new system. They are most likely very comfortable with the reports they receive and the manner in which they execute them. And changing an "old" process that people have done for years, every day, the same way, is very tough. Thus, here is what I propose:

1. Talk to the users and find out which are the Top 5 most critical reports (daily, weekly, monthly)

2. Recreate these reports in the new system
- this is important to confirm to the users that the new system produces the same results as the old one

3. Talk to the users and find out what are the biggest "drawbacks/pains" for using the current reporting system

4. Create new reports to address these issues
- especially look to replace manual data merge processes
- i.e. a user runs two reports and combines them manually in Excel
- you should build the final composite report in the new system (emphasize --> one click and you are done)

5. Talk to your executives and find out which high-level (re: dashboard) summaries would they like to see

6. Create dashboards for these summaries using the new system

Eventually, the users will see:

a. the data is the same and accurate in both systems
b. the new system is more capable / usable than the old
c. there are additional insights to be gained from the new system

Good luck...

[ngalemmo]

I hate to say this, but it should have started 18 months ago. What was the genesis of the project? Wasn't the business driving it? Or, was this an IT project to 'fix' a perceived problem with users reporting off a replicant?

Historically, IT departments get roundly critisized by the business side because they tended to drive change in systems when the need for change was not perceived as needed by the business. Have you ever wondered why large organizations are littered with old redundant systems that should have been retired years (or even decades) ago? The biggest mistake in data warehousing is to build a "Field of Dreams". That is, a DW without sponsorship by the business. If you build it, chances are nobody is going to come.

At this point, you have an uphill climb. Elmorejr has some good suggestions. If you had engaged the business and user community from the start and had them actively involved in the project, they would be the ones breathing down your neck asking when it will be ready. Put the matix and diagrams in your drawer and instead, show the users how to get their current reports out of the new system (and I do, literally, mean show them).

[mselway]

Well ngalemmo you seem to have run off on a personal fantasy.
Perhaps you should wait for an answer to your first question before sounding off about your own issues.

Thank you elmorejr for constructive suggestions.

[ngalemmo]

Ok. whatever.

[BoxesAndLines]

Go to the CIO and tell them you have redundant reporting systems that is costing an arm and a leg to the company. That should initiate a project to migrate all the reports to the new warehouse. You should also be prepared for the alternative.

[Sponsored content]