The quick answer to this question is NOW!
However for most organisations, performing a data quality review can be a long-winded exercise.
From my experience, following are some perfect opportunities to not only perform a data quality review, but also to resolve any data quality issues:
1. When planning to implement a new system and migrate existing data to it
Why? Because you don’t want to be implementing a new system and transferring all the existing data issues to it. This is generally the one time where it can be easily justified as to why you need to get your data quality right.
2. When you need to share data with trading partners
Why? Because you don’t want your organisation to look silly by providing information that is out of date or incorrect as the resulting issues can be costly. Imagine shipments being sent to the wrong customer address or your manufacturing partner has just made 10 times more parts than you require for your equipment, due to incorrect data that your organisation provided.
3. When you need reliable information to perform analytics or make decisions
Why? Because you want to be making decisions using information that is accurate! What’s the point of making a decision on which existing product line to invest in, when your inventory transaction data is incorrect?
4. When you need assurance prior to an audit
Why? Because failing an audit can be a costly exercise, ranging from warnings to loss of licence to operate.
Why? Because your organisation should be operating with accurate and reliable information at all times. You cannot afford not to, especially when the stakes are high. It also helps your organisation respond to an event much faster.
These are just some examples that I have come across in my career. What other instances or examples have you come across?