Exactly correct data for example is a requirement for financial and regulatory reporting - but dircetionally correct data is also useful, even if it is not perfect. Especially in marketing and sales directionally correct data is important and helpful for building predictive models. If you use House of Quality methodology to determine what level of quality is required for which data this approach means continous improvement.
The chosen strategy has also implications for the data strategy, like consolidated data vs. federated data or historical data vs. real-time-data. Consolidated Data e.g. provides cost efficiency and consistency - and it requires a lot of up-front investment to make it happen. Federated Data, in contrast, provides agility and flexibility in how we look at data. Here, you have to pay a higher cost at query time in order to avoid up-front costs for building a consolidated data environment.
If you combine these, you can create a lifecycle management. Together this data provides a more robust platform for reliability and sustainability and more efficient work with data labs with federated data.