Do I need to clean or prepare my data before using iDBQuery?

A lot of analytics projects stall on 'we need to clean the data first'. iDBQuery is built to work with real-world data as it actually exists.

• Connect as-is. Point iDBQuery at your production database, spreadsheets or files without reshaping them into a perfect model.
• The semantic layer bridges the mess. It maps obscure column names, internal codes and abbreviations to plain business meaning, so a table called `cust_dim` with a `sts` flag still answers questions about customers and status.
• It handles imperfections at query time. Nulls, duplicates, mixed formats and cryptic keys are dealt with in the generated SQL rather than in a preprocessing pipeline.

That does not mean data quality stops mattering — garbage in still limits what any tool can conclude. But iDBQuery lets you start immediately and even use it to find the problems: you can ask it to surface duplicate records, missing values or figures that disagree between systems. For example, before a month-end close you might ask 'show invoices with no matching customer', turning data cleanup itself into a plain-language task. So rather than cleaning first and analysing later, you connect, ask, and clean up as you go.