How do you manage change and adoption of the new BI system after data migration?

As part of preparation, engage data owners and users to agree taxonomy between sources and destination. Does "exchange rate" mean inter-bank lending as reported in WSJ or is it what I will see at the Heathrow airport? Getting this agreement will be essential for the transform process. As part of planning, include any documentation and test results which will be required for audit requirements. Can you adequately prove to FDA, SEC, IRS that the target data is valid?

Adding a data health check in a data load process is a critical step to ensure the accuracy and integrity of the data being loaded. Here are some steps you can take to add a data health check to your data load process: Create a data quality check script: Once you have defined your data quality metrics, create a script that checks your data for these metrics. Integrate the data quality check into your data load process: Once you have created the data quality check script, integrate it into your data load process. This can be done by running the script before the data is loaded into your system. If the script identifies any data quality issues, it can either halt the data load process or flag the issues for further investigation.

Three high-level metrics would be great to track and manage. Time-to-Data: refers to the time it takes to retrieve and load data from various sources into a data mart in the data warehouse. This metric includes the time needed to identify relevant data sources, extract and transform the data, and load it into the data mart. Time-to-Report: refers to the time it takes to create, modify, or maintain a report or BI application. This metric includes the time needed to design and develop the report and the time needed to gather and analyze the data, create visualizations, and perform any necessary testing. Time-to-Action: measures the time taken to make a decision or take action based on information provided by a report or BI application.

Adoption = Value. Best example I've seen? A 160 year old insurance company implementing a BI solution across five lines of business. If you didn't access the system within three (3) days of your training, the Solution Support Team swung by your desk to see why. Lack of engagement ranged from "I was busy" to "I couldn't log in" to “It doesn’t meet my needs." No "write-ups" or giving you a hard time. The mission was to ensure you and the company got value from the MASSIVE investment. Each were highly trained, empathetic, knew the solution and YOUR area of the business. Rarely would they fail to address ALL your concerns on the spot and get you productive. The team sought out gaps in the solution to drive the next round of improvements.

In my experience, if you are a smaller business, the temptation is to cut corners and save costs. However, the data migration step is the most important one, where if you are not careful, you can introduce a lot of noise into your dataset. This will then have a long term impact on the resource as a whole going forward. It is also impossible to revert, once this step is over. Having the right consultant/specialist to support this is a good investment and will save you many headaches later down the road.