How to Clean Data for an ERP Implementation

Access to current and historical data is one of the top benefits of an ERP system. But if that data in inaccurate or untrustworthy, it can compromise quite a bit of the worth of the system — and frustrate the users to no end. It’s no secret that a good data cleaning strategy must be in place long before switching over to a new ERP system, but organizations continue to persist in the idea that they can migrate everything they have over to their software and the software will somehow magically scrub and standardize it. That’s just not the case. And if they’re counting on their ERP vendors to help them, they’re in for a real surprise. Data migration is such an onerous and extensive task that many vendors simply leave it up to the implementing company to figure out. Suffice it to say, organizations need to get their data strategy set and begin working on it long before the cutover date.

Following are some basic tips on how to approach cleaning historical data for an ERP implementation:

1. Decide what you really, really need. An enterprise software implementation is a great opportunity to “clean house.” Organizations tend to hoard data, believing that someone, somewhere down the line just might need that one nugget of information and if it can’t be found — disaster will strike. The problem with this attitude is a. there’s a lot that no one will ever need and b. dragging it all over to a new software system will take time, money and effort that can be better spent elsewhere. A person or team with the proper authority needs to be tasked with working with different departments to determine what’s truly necessary to keep and what can be jettisoned. Even the IRS doesn’t make you records for more than a few years!

2. Put some elbow grease into it. Now that you’ve decided what you need, it’s time to really get cracking on the cleaning part. Don’t make the mistake of giving this task to the IT department; instead, ensure they work together with the functional users in each department to standardize the records, validate the information, de-dupe, input missing data, etc. Bear in mind that the specifics of data cleaning (e.g., in-system vs. Excel) will likely need to be determined with the help of a third-party consultant, especially if your organization is dealing with industry-specific regulatory requirements.

3. Migrate and test. Discover if you’re on the right track with your data by migrating it to the test environments and taking a looksee. I’ve made this sound much easier than it actually is. In reality, this will take quite some time and typically require the development of a lot of unique code, so don’t make the mistake of leaving it until the last minute. The better this is managed, the smoother your transition to your new ERP system will go.

People often compare data migration to moving houses. You’ve got to decide what to bring with you, clean everything up that you want and make sure it all fits in the new structure accordingly. This analogy is accurate in a way. You can probably get a closer read on the situation by adding to it that your part of a family that hasn’t thrown anything away in years and doesn’t know how to make decisions about what to keep or toss. Oh, and the house you’re moving into is incredibly strict about what you can bring in and if you bring something wrong, it will lock down and throw you out.