Decentralise with data mesh
Suppose your organisation is a bank and you offer a range of services: you provide current accounts, mortgages, and have a wealth management service for the luxury segment. All data for these three different services is stored in a central data warehouse. But who can understand and combine this data into valuable insights?
For years, organisations strived for a single database where all data is stored together. In practice, we often see that this centralised approach creates an excess of data flows. Moreover, a centralised data team does not necessarily have enough domain knowledge to convert all that data into insights, and make it available for valuable applications.
Opportunity
Data mesh is still a promise, not a ready-made solution. However, if your organisation has large amounts of data, and wants to be data-driven, the step to data mesh is a great opportunity to keep in mind. With the data mesh idea, responsibility for structuring that data isn’t with a central team. Different product teams manage parts of the customer journey or a customer application. You don’t just throw all the data into one barrel. You make connections between the different teams so that all the different data sources and collection methods can coexist, but with data mesh you have the certainty that you have drawn the right lines to let the data flow from one team to another. As a result, data & intelligence, which was previously exclusively an IT matter, is a responsibility and resource that’s shared across the business.