Free - No signup - Source data stays in your browser

Prepare CSV for Power BI.

Review headers, Power Query types, locale, and one optional business key across the complete file. Export normalized UTF-8 CSV plus reproducible M - without uploading the source.

Power BI CSV preparation means controlling encoding, headers, types, culture, and key quality before import. This local workspace validates every row, preserves identifier text, and produces both the reviewed CSV and an explicit Power Query M starter.

01 - How it works

Replace inference with a reviewed contract.

Power Query can infer a useful preview and still choose the wrong type for a later row, an identifier, or a locale-sensitive value. Review the complete file first, then carry the decisions into the import.

i. load

Open CSV or Excel

Drop a file, paste CSV, or use the imperfect sample. Multi-sheet workbooks stay local while you choose the intended sheet.

ii. review

Confirm names and types

Keep leading-zero identifiers as text, choose culture deliberately, edit output headers, and optionally test one business key.

iii. hand off

Export CSV and M

Resolve written findings, download the complete normalized CSV, and copy or download the matching Power Query starter.

02 - Why this workflow

A Power BI import with evidence.

The goal is not to imitate Power BI in a browser. It is to catch the silent choices that change identifiers, dates, decimals, joins, and refresh behavior before the destination owns the data.

  • 01

    Full-file inference

    Suggestions inspect every nonblank value, never only the first 200 rows. Mixed values remain text until you make a deliberate type choice.

  • 02

    Culture-aware checks

    Select en-US, en-GB, de-DE, or fr-FR and revalidate every typed value. Ambiguous slash dates are never assigned a hidden culture.

  • 03

    Identifier protection

    ID-shaped columns and digit strings with meaningful leading zeros default to text so import preparation does not erase identity.

  • 04

    Reproducible M

    The generated query fixes comma delimiter, UTF-8, quote handling, promoted headers, complete reviewed types, and culture in readable Power Query code.

"A preview is a sample. A reviewed schema is a decision."
csvtodashboard.com
Scope and handoff

Know what this preparation does not create.

The workspace creates a local CSV and Power Query starter. It does not create a PBIX, semantic model, relationship, measure, report, publication, gateway, credential, refresh schedule, or guarantee of downstream correctness.

  • Data stays unchanged

    Header edits affect the exported header row. Source cell text and row order are preserved; the tool does not coerce values merely because you selected a Power Query type.

  • Validation is destination-shaped

    Type and key checks expose import hazards. They do not replace representative visual totals, relationship tests, refresh tests, or business-owner approval inside Power BI.

  • The path is a placeholder

    The M query shows exactly which path it will read. Change it deliberately after placing the normalized CSV where Power BI should find it.

03 - FAQ

Power BI CSV questions.

How do I prepare a CSV for Power BI?
Review the delimiter and encoding, preserve identifier columns as text, choose a deliberate culture, validate every value against explicit Power Query types, check any business key, then import a representative normalized CSV before relying on refresh.
Microsoft documents that the Text/CSV connector automatically infers UTF-8 only when a UTF-8 BOM is present. The generated M query also declares encoding 65001, so both manual and scripted handoffs are explicit.
Values such as 00101 are identifiers, not quantities. Assigning a number type can remove meaningful leading zeros and change joins, labels, and exports, so identifier-shaped columns default to text.
No. Slash dates are culture-dependent, so they stay text until you choose a culture and deliberately select Date or Date/time. ISO dates can be suggested without making that hidden locale choice.
It reads an explicit local path with File.Contents, parses comma-delimited UTF-8 with CSV quote handling, promotes headers, and applies every reviewed column type using the selected culture.
No. It creates a normalized CSV, a Power Query M starter, and an aggregate readiness report. You still test and model the data in Power BI Desktop or the Power BI service.
You can select one business-key column to find blanks and repeated normalized values across the complete file. The tool does not infer database constraints, relationships, measures, or dimensions.
No. Parsing, type review, validation, preview, query generation, and downloads happen in the browser tab. A data handoff occurs only when you later import or connect from Power BI.