Recently, one Sunday afternoon, my wife told me her purchased airline reservation for Monday had disappeared. A few days ago it was there, then suddenly it was gone. Apparently she was not alone. Complaints were coming in online. Support was overwhelmed. In fact, Twitter seemed to be the only way to get a quick response, so people were sending their messages and details that way!
Now, this is a great brand that I’m talking about, what I would call the chic brand in the airline industry. They had just experienced a brand-killing event. No one wants to fly an airline where reservations disappear. But the next part stunned me. When my wife finally talked to a manager at the airline, it turned out they expected this would happen! They had just migrated to a new reservation system and another airline that had done this same migration had experienced issues for a few weeks. So the airline not only expected to have similar issues, they scheduled fewer planes to fly in the upcoming weeks!
This airline could have avoided this brand-killing event if they had just watched for missing data during the data migration process. And today, this can be done. They just didn’t know. Thousands of companies today use Informatica not just for data warehousing or operational integration, but for data migration as well. Many of these customers don’t just implement data integration. They also use data migration as a starting point for data governance. That’s partly because data migration can change a lot of data; sometimes even lose a lot of data like reservations. Data governance is meant to prevent that.
More recently, some of these customers have been using a new type of product for data governance called Informatica Data Validation Option, or DVO, to verify that all the data was moved properly before going live, and also to check on a daily or weekly basis that the data used in production is accurate. For migrations, it’s pretty straightforward. DVO generates the tests you need to compare data in an old system to data in the new system. Instead of 1-10% of the data being tested in a typical data migration, 100% of the data is now tested. And this new way of testing takes one tenth the original effort because these tests used to be hand coded. DVO is used for migration projects, for general testing and also for validating any production data. Why, because bad data or missing data can lead to costly mistakes like what happened to this airline. DVO can catch issues with data created by bugs or system failures before they impact the business.
So when migrating to new applications, don’t forget about to verify 100% of the data. Otherwise, you might find out the hard way how important that data really is to your brand and your bottom line.