Operational Reporting

Hello Everyone,

We have a Data Warehouse built on Kimabll Architecture(Business Process and Grain Driven). Which is working fine and serving the purpose for Performance Management and Analytical Reporting. All data comes from a home grown source system which performance various funcations (Order Entry and Management, Billing to the customers, product life cycle management etc) Now this home grown system may be replaced with various out of the box solutions. So we may have products from few vendors to replace the home grown system. My question is around operational reporting. Is data warehouse the place to do operational reporting but again data warehouse is process centric and at any point in time we may not have all the subject areas in the data warehouse needed for operational reporting. In this case should we be build an Operational Data Store to do the operational reporting. Yours comments on this are appreciated.

It depends. Sometimes the data warehouse, even if it had all the subject areas, is not suitable for all operational reports.

Depending on your definition of operational reports, it may require access to transient data elements that either don't exist in the DW or are not timely enough to be useful. (For example a data warehouse may not have orders 'on hold' pending acceptance or rejection, but rather only orders after the accept/reject decision has been made). There is usually little or no business value (from a strategic or tactical perspective) to capture such information. Whereas, an operational data store is a different animal for a different purpose. In some cases, an ODS is nothing more than a replicant of the operational system put out there to offload the query burden.

Thanks for the reply. This is so true in our case. Our data warehouse has processes which we decided to bring in the data warehouse. For operational reporting the requirements are very broad and for sure they are out of scope of the data warehouse. In this case ODS makes perfect sense. Which data modeling approach is better for designing ODS, Data Vault or Relational. Please, comment.

I would go for a near-3NF relational model. I cases where I have done this, 'master data' and status information was all current state, so other that transactional data, there was no need for a historical perspective beyond what the operational system itself provided.

Operational reporting is designed to support the detailed day-to-day activities of the corporation at the transaction level. In operational reporting, detail is much more important than summary. In fact, in operational reporting summary information is often irrelevant. Examples of operational reporting include bank teller end-of-day window balancing reports, daily account audits and adjustments, daily production records, flight-by-flight traveler logs and transaction logs.
Data Entry India