Glossary
Schema drift
Schema drift is when the structure of a dataset changes over time, fields get added, renamed or removed, without the systems consuming it being updated to match.
Schema drift is when the structure of a dataset changes over time, fields get added, renamed or removed, without the systems consuming it being updated to match. It matters in e-commerce data because a pricing engine or dashboard built against one schema will break or silently misread data if a field disappears or changes meaning. Well-maintained data feeds handle this by versioning the schema, making changes additive rather than destructive, and announcing removals well ahead of time rather than dropping a field without notice. Buyers of ongoing data feeds should ask how a provider handles schema changes before building anything that depends on a fixed structure.
Related terms
Where this shows up
Need this field in your own feed?
Tell us which stores and fields you need. We reply within one business day with a scope and a price.