Managing Your AMS Implementation: Tip #4 Prep Your Data

Planning for an AMS implementation can be a huge undertaking and involve a great many staffers in your organization. Here’s a quick tip that can help you plan ahead.

magnifying glass data streamTip #4: Prep Your Data

One of the most detailed and time-consuming tasks in any AMS implementation is making sure the data is prepped and ready to migrate. Here are some steps to help guide you:

Decide what should be migrated: big buckets and then field level

  • How easily can your data be exported?
    • Do you have an export capability built in to your current system? If not, can you talk to your current vendor and figure out how to export the data?
  • Validate your data import budget
    • Often, the budget for the data import to the new system isn’t adequate. Make sure you’ve budgeted enough time and money to not only import the data, but to verify that it’s all correct and account for any subsequent clean-ups and re-imports that might be necessary. If you’re data’s really clean, one import pass may do it. If not, it may take several.
  • Run data integrity checks
    • Does your old system have data validation? If not, you may have some messy data in some key fields.
    • For example, a ‘phone’ field from an older database we once worked with was actually an alphanumeric field that could hold ANYTHING. So, we saw values like “(703) 123-4567” and “7031234567” and “703-123-4567” which aren’t too bad to work with. BUT we also saw values like “Call 703-123-4567 first but if no answer try 4566 instead” or “Need to verify whether we should dial into main line or extension 344” that caused issues with the import.
    • Also, if an older database was in use for many years and several generations of staff, one field may be holding different data. If so, this field must be split up or the rules must be communicated clearly to the vendor. For example, one older database had an alphanumeric field that for a few years had been used to hold an annual revenue number (e.g., 1,000,000,000) but in the later years, staff had changed to putting in a code that represented their revenue size (e.g., 3 meant a medium sized revenue company).  With no info about this shift, the import of this field obviously would obviously go wrong.
  • Perform data clean-up before the import
    • Don’t wait until you’ve imported the data to the new system to clean it up. Check for duplicate entries, partial entries, incorrect information and make all the changes you need to beforehand.

The data’s prepped, your team’s ready. What’s next? Tip #5: Gather Samples

 

Joanna Pineda

About Joanna Pineda

Joanna’s business card reads CEO/Chief Troublemaker for a reason. She relishes a challenge and introduces change wherever she goes. She knows anything is possible and that clients come to Matrix to hear "Yes", not "No." Matrix is purple because of Joanna. Staff like to call her JP.

Leave a Reply

Your email address will not be published. Required fields are marked *