Decide where the work belongs
Map the relationships that depend on identity
List the joins, updates, or references that use the imported field. For a bounded sample, compare the exact source and stored strings and record failed matches. Distinguish observed relationship damage from workflows that could be affected but have not yet been checked.
Group susceptible input shapes
Compare long digit strings, values with leading zeros, and short identifiers that survive unchanged. Group examples by producer and mapped field. Estimate reach from the relevant import population rather than assuming all numeric columns or every file exported by the same tool has the defect.
Assess a safe correction path
Determine whether original identifiers remain available and whether affected records can be reconciled without guessing. Check whether a text type can be selected before import. Include the effort of reviewing ambiguous matches, since simply repeating the import may create more unmatched or duplicate records.
What to carry forward
Set priority using confirmed identity damage, recovery confidence, affected workflows, and prevention available at preview. Assign a reviewer for ambiguous reconciliation. Escalate when altered identifiers are silently accepted into an update path; defer only with a clear scope note and a reliable way to preserve the source strings.
Technical background: PostgreSQL documentation.
Keep the decision with the work.
Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.
Create an account See the product workflow