For years, digital growth was often associated with collecting more: more events, more fields, more history and more copies. Storage was affordable, and future analytical value seemed like a sufficient reason to keep information.
That assumption is changing. Larger datasets can mean more exposure, more maintenance and more uncertainty about why information is still held. Organisations that can identify what they need, and confidently retire what they do not, may move faster and manage risk more effectively.
More data is not always more value
Every field should support a real activity. If a team cannot explain why it collects an attribute, how it is used or when it stops being needed, the data may create work without creating meaningful value.
Unnecessary information can complicate product design, increase the impact of an incident and make access, correction, retention and deletion requests harder to fulfil. It can also reduce analytical quality when old, duplicate or poorly understood records are treated as reliable signals.
The advantage may come not from having the largest data estate, but from knowing which information earns its place.
The cost of keeping data
Data has a lifecycle cost beyond storage. Teams need to secure it, manage access, maintain integrations, document purposes and respond when information changes. Copies may remain in exports, backups, test environments or third-party platforms long after the original process has ended.
Retention decisions are not simply technical housekeeping. They connect business need, legal obligations, records management and privacy risk. A clear schedule should explain what is retained, why it is retained, which event starts the retention period and what happens when it ends.
Minimisation improves quality
Reducing unnecessary collection can make processes easier to understand. Shorter forms can be easier for customers to complete. Smaller datasets can be easier to validate. Fewer redundant copies can simplify access control and improve confidence in the data teams actually use.
Minimisation also asks teams to separate a genuine requirement from habit. If an existing form requests a field only because it has always been there, a review may reveal that the field is no longer needed or can be replaced with a less intrusive signal.
Put minimisation into the workflow
Make data necessity a normal design question when teams create a product feature, a customer journey, a new report or a vendor integration. Ask what decision each field supports, whether a less detailed value would work, who needs access and how long the information is useful.
Classification and discovery can support the review by showing teams where certain categories of data live and which processes use them. RoPA and retention records then provide a place to document purpose, ownership and lifecycle expectations.
Measure what you no longer need
Useful measures include the number of fields removed from a journey, datasets assigned owners and retention rules, stale records disposed of, duplicate stores consolidated and processes with a documented purpose review. These measures should be interpreted in context; fewer records alone do not prove that a process is well governed.
Start with one high-volume process. Map the data collected, involve the people who use it, identify fields with no current purpose and test a smaller version of the journey. Record both the operational benefit and any new control needs.
Data minimisation is not about making organisations data-poor. It is about making collection, use and retention deliberate. That discipline can become a source of efficiency, trust and resilience.
