Data Integrity Built In
Data quality problems are insidious. They often start small — an orphaned record here, a broken reference there — and compound over time until the data your business depends on can no longer be trusted.
Most of these problems have a single cause: data was changed in one place without accounting for its connections to other data.
The Integrity Problem
Consider an order that references a customer. If the customer is deleted without handling the orders that belong to them, those orders now reference a customer that doesn't exist. The order detail page breaks. Reports produce wrong numbers. Support can't find the customer for an order they're trying to resolve.
This is a referential integrity failure — and it's one of the most common and damaging data quality issues in business applications.
Automatic Consistency Checks
In Swifty, relationships between object types carry automatic integrity constraints. When you define that an Order relates to a Customer, the platform knows that a Customer referenced by Orders cannot simply be deleted without resolution.
The platform enforces this automatically. If you try to delete a Customer that has associated Orders, you're prompted to resolve the dependency first — reassign the orders, archive the customer, or delete the orders along with the customer, depending on what your business rules dictate.
You don't configure this constraint explicitly. It's a consequence of defining the relationship.
Cascades and Protections
Beyond deletion protection, relationships can be configured with cascade behaviors. Archiving a project can automatically archive all tasks beneath it. Cancelling an order can trigger a status change on related fulfillment records.
These cascades run inside the platform's integrity layer — they're not loose automations that might or might not fire. They're structural behaviors tied to the relationship definition itself.
Validation Before Save
Integrity checks also run before records are saved, not just before they're deleted. Required relation fields are validated. Format checks run on field values. Cross-field rules can be defined to catch inconsistencies before they're committed.
The principle is consistent: catch data problems at the point of entry, not downstream when they've already propagated.
Trustworthy Data
The compounding benefit of built-in integrity is data you can trust. Reports that aggregate across relationships produce correct numbers because the relationships are sound. Detail pages that display related data load correctly because the references are valid.
Trustworthy data is the foundation of every business decision made from the platform. Data integrity built in means that foundation is solid — not because people remember to check, but because the platform makes inconsistency structurally difficult.