What steps would you take to respond to a data quality issue you discover?

• A. Ignore the issue until it worsens.
• B. Blame the data consumer.
• C. Re-run the process without investigation.
• D. Reproduce issue, trace data lineage, assess impact, inform team, document, implement fix and verify.

Answer: (D)

When you discover a data quality issue, respond with a structured, end-to-end approach that not only fixes the problem but also preserves traceability. Start by reproducing the issue in a controlled setting to confirm exactly what is wrong and under what conditions, so you’re observing the real symptom rather than guessing. Then trace the data lineage to map the data from source to destination, which helps pinpoint where the fault likely originated and how it spread through the pipeline. Next, assess the impact to understand how widespread or critical the problem is, who needs to be alerted, and what level of remediation is required. Inform the right team members so everyone can coordinate the response and avoid duplicating effort. Document what happened, the steps taken, and the evidence gathered so there’s a clear, auditable record for future incidents. Finally, implement a fix, verify the resolution by re-validating data quality (and re-running relevant checks or end-to-end tests), and confirm that the data meets quality criteria again. This sequence reduces risk, improves trust in data, and helps prevent recurrence. Ignoring the issue, blaming others, or simply re-running the process without investigation won’t address root causes or prevent repeated problems.