Data Decay Definition
Data decay is the gradual loss of accuracy in stored data as the real-world facts it describes change over time. A record that was correct when it was entered, such as a customer's address, a supplier's contact person, or a product's price, becomes outdated when the facts change but the record does not. Data decay does not involve errors in entry or storage; the data simply stops matching reality.
What causes data decay?
Most data decay comes from ordinary changes in people, organizations, and products:
- People change: customers move, change email addresses or phone numbers, get married and change names, or leave their jobs.
- Organizations change: companies merge, rebrand, relocate, close, or restructure, which affects customer and supplier records.
- Products and prices change: items are discontinued, specifications are updated, and prices or availability shift.
- Rules and reference values change: tax rates, regulations, country codes, and industry classifications are revised from time to time.
Data can also decay when a change is recorded in one system but never reaches the others, leaving outdated copies behind.
Which types of data decay fastest?
Contact data decays fastest, especially in business-to-business (B2B) settings, where job changes and company restructuring are frequent. Customer and supplier master data are therefore among the most affected. Data that describes stable facts, such as a product's physical dimensions or a country's name, changes rarely and decays slowly.
Why does data decay matter?
Outdated data leads to practical problems that are often hard to trace back to their source. Examples include invoices and deliveries sent to old addresses, marketing emails that bounce, orders placed with suppliers using expired terms, and reports built on information that no longer applies. Because decay happens gradually, the overall data quality of a system can drop significantly without anyone noticing a single moment of failure.
How can organizations reduce data decay?
Data decay cannot be stopped entirely, but it can be managed. Common practices include:
- Regular review: checking data for records that have not been updated for a long time or show signs of being outdated, such as bounced emails.
- Ongoing cleanup: correcting, updating, or archiving outdated records through data cleansing.
- Clear responsibility: assigning specific people to maintain specific data, so updates do not fall through the cracks.
- Keeping systems in sync: using a master data management (MDM) approach so that an update made in one place reaches every connected system.
- Update at the source: letting customers and suppliers update their own details, for example through account settings or a supplier portal.
Is data decay the same as data corruption?
No. Data corruption, sometimes called bit rot, is technical damage to stored files or records caused by hardware failures, software bugs, or transmission errors, so the data itself is broken. With data decay, the data is intact and readable but no longer accurate because the world it describes has changed.