Change is a constant in life. And when it comes to your data, even more so. Given enough time, migrating your data to new systems and platforms will be a fact of life for most businesses. Whether it involves a corporate merger, a new application vendor, or other reasons, data migration is one of those predictable “stress points” that can put your contact data assets at risk without the right strategy.
According to a recent post by Dylan Jones on Data Quality Pro, data quality issues are one of the key reasons for the high failure rate of data migration projects. He cites a recent survey showing that 84% of these projects run over time and/or budget – and in his view, an important part of avoiding this involves advance planning and modeling.
Data quality best practices for contact data migration
From our perspective, there are at least four best practices you should consider for preserving contact data quality during a data migration process:
Measure twice, cut once. Jones describes the use of what he calls landscape analysis, or a “migration simulation” in plain English, to anticipate problems before the migration begins in earnest. This involves testing a subset of your database against planned conversion rules and protocols, to help ensure that the results are likely to go as planned.
Validate addresses before conversion. The term “garbage in, garbage out” applies here, with clean data being an important factor on the front end of the migration process.
Validate addresses after conversion. What is the one thing that is worse than bad data? Reformatting data to make it even worse. Bad things can happen when old data goes into new fields, and if John Smith at 123 Mayberry Street becomes “John” at “Smith 123 Mayberry” in the new database, his value as a contact can go completely out the window. Never blindly trust converted data without a validation and review step to flag bad contact data in real time.
It ain’t over until it’s over. Yogi Berra’s famous baseball saying applies equally to data migration, because you aren’t finished with data quality when the initial conversion is done. Contact information is a perishable resource that goes stale over time, as people come, go, and change jobs and addresses. This means that your contact data migration isn’t really finished until you have implemented and tested an infrastructure for ongoing contact data validation and database cleaning.
Data migration: Not just a technical problem
According to a recent Infosys white paper, treating data migration as an “IT problem” is often a fatal mistake in terms of data quality – in their words, “Business has to not just care about data migration, but command it.” Put another way, no one can sweat the details on data quality like the stakeholders who will be using this data in the long run.
This raises one more important issue: a major data migration might also be the time to start thinking about a more formal data governance strategy, if one isn’t in place already. We’ve discussed this issue on our blog before, and it is particularly relevant here: major changes such as a data migration can often serve as a catalyst to build professional data expertise at a business-wide level. Either way, putting data quality front and center is one of the most important factors in creating a smooth transition to a new environment.
Have any additional questions on the important role data quality plays in the data migration process? Contact us and we will be happy to answer your questions.
