Data Looks Wrong

Use this runbook if a figure in the Yorkshire Vitality Suite dashboards looks incorrect, is missing, or appears not to have been updated.

---

Step 1 - Check when the dataset was last loaded

Connect to the database (see Query the database) and run:

SELECT dataset_code, extraction_status, rows_extracted, rows_loaded, loaded_at, error_message
FROM dataset_metadata
ORDER BY created_at DESC
LIMIT 30;

Look for the dataset in question and check:

• Was the most recent run a success or a failure?
• Does rows_loaded match rows_extracted? A mismatch suggests some rows were rejected.
• When was loaded_at? If it is several months ago, the flow may have stopped running.

---

Step 2 - Check the indicator values directly

Run a query against the indicator table:

SELECT indicator_id, indicator_name, lad_name, reference_period, value, updated_at
FROM indicator
WHERE indicator_id = 'your_indicator_id'
ORDER BY reference_period DESC, lad_name;

Replace your_indicator_id with the relevant indicator (e.g. employment_rate, life_expectancy_male). This shows you exactly what is stored in the database.

---

Step 3 - Compare with the source

If the value in the database does not match the source publication, the data may have been loaded incorrectly.

Symptom	Likely cause
Value is much older than expected	The flow has not run recently - check dataset_metadata
Value is present but obviously wrong	The source may have revised the figure - compare directly with the API or publication
Value is missing entirely	The row was not loaded - check the rows_extracted vs rows_loaded discrepancy
Dashboard shows no data	Power BI may be using a cached version - try refreshing the report

---

Step 4 - Re-run the flow

If the data is simply out of date, re-run the relevant flow manually. See Re-run a flow manually.

After the flow completes, check dataset_metadata again to confirm rows_loaded is as expected, then refresh the Power BI report.

---