How do I investigate an anomaly or unexpected number in my data?

Question

iDBQuery · Accepted Answer

To investigate an anomaly in iDBQuery, ask why the number looks off and let the autonomous Analyst agent dig in. It queries your live data across sources, isolates what changed — a segment, a period, a bad record — and returns a written, cited finding pointing to the exact rows behind the spike or drop. When a number looks wrong — revenue dipped, costs jumped, a metric flatlined — the usual response is hours of manual slicing. iDBQuery turns that into a question.

- Just ask — "Why did refunds spike in March?" iDBQuery writes the queries, compares periods, and breaks the figure down by the dimensions that matter.
- Hand it to the Analyst — the autonomous agent investigates on its own: it forms hypotheses, runs follow-up queries across every connected source, and narrows to the root cause instead of stopping at the first chart.
- Get a cited finding — it returns a plain-language explanation with every claim traced to source rows, so you can confirm the cause rather than take its word.
- Cross-source by default — an anomaly explained by data in another system (a pricing change in one database, a campaign in a spreadsheet) is caught because everything answers as one model.

This is the difference between a tool that shows you a number and one that explains it. iDBQuery doesn't just surface the anomaly — it tells you which transactions, customers, or records drove it, with a trail you can audit. If the cause turns out to be a data-quality issue, the citations point straight at the offending rows.

كيف أحقّق في حالة شاذّة أو رقم غير متوقّع في بياناتي؟

Related questions