Snowflake CSV vs CSV for Excel: Which Should You Choose?

How-to basics · Updated November 20, 2026

Snowflake CSV is the standard comma-separated export, while CSV for Excel is a Snowflake download option intended to help Microsoft Excel recognize the file’s UTF-8 text correctly, typically by including a UTF-8 byte-order mark (BOM). Choose CSV for Excel when you plan to open the download directly in Excel and it contains non-English characters; choose standard CSV for a general-purpose file or another data tool. Neither option is an Excel workbook.

What the Two Snowflake Options Mean

CSV stands for comma-separated values. It is a plain-text format for rows and columns, not a spreadsheet file with formatting, formulas, multiple worksheets, charts, or Excel-specific features. A CSV can be read by Excel, databases, scripts, and many other applications, but each application decides how to interpret the text.

In Snowsight, Snowflake’s web interface, CSV and CSV for Excel can appear as separate choices when downloading query results. The Excel-labeled choice is meant to make the downloaded text easier for Excel to identify as UTF-8. UTF-8 can represent characters from many writing systems, including accented letters, non-Latin scripts, and symbols. A BOM is a short marker at the start of a text file that can help an application identify its encoding. It is not a visible character in the spreadsheet.

The exact choices and controls you see can depend on the Snowflake interface and the way you are exporting. The important distinction is that CSV for Excel is an Excel-oriented download choice, not a different kind of Snowflake table or a replacement for the CSV file format used in SQL.

The Practical Difference: Text Encoding

For ordinary English-language values, the two downloads may look identical when opened. Their rows, columns, and comma-separated structure are intended to carry the same query results. The Excel-specific option’s useful difference is how the file signals text encoding, rather than a change to the meaning of the database data.

For example, suppose a query returns the name José, the city 東京, and the symbol €. If Excel guesses the wrong encoding when a plain UTF-8 CSV is opened directly, some characters may appear as question marks or garbled text. A BOM can make UTF-8 easier for Excel to detect. It does not translate text, repair characters that were already corrupted in the source, or change the query’s values.

A BOM also does not solve every CSV-opening issue. It does not specify which character separates columns, tell Excel that a column contains dates rather than text, or preserve display formatting. Those are separate interpretation choices made by Excel or by the import settings.

How Excel Opens a Snowflake CSV

On Windows, double-clicking a CSV commonly opens it in Excel using the computer’s regional settings and Excel’s automatic interpretation. This is convenient, but it can produce surprising results: a long identifier may be displayed in scientific notation, a value with leading zeroes may lose them, or a text value that resembles a date may be converted into a date. The CSV itself does not store a column type that prevents those conversions.

For more control in Microsoft 365 for Windows, open Excel and use Data > From Text/CSV, select the file, and review the preview before loading it. In the import experience, check the delimiter and encoding, then use the available transformation options to set sensitive columns to Text when needed. This is especially useful for account numbers, postal codes, product codes, and identifiers with leading zeroes. Excel for Mac has text-import and Power Query options too, but labels and availability can differ by release; follow the import controls shown in your Mac version rather than assuming the Windows dialog is identical.

Importing is usually preferable to double-clicking when the file contains international text or values that must remain exact. It gives you a chance to inspect how columns are split and typed before the data enters the worksheet. If you only need to view a small, simple export, opening CSV for Excel directly may be quicker.

When to Choose Each Download

Pick based on what will consume the file, not on which name sounds more complete. A useful rule of thumb is:

  • Choose CSV for Excel when a person will open the download directly in Excel, especially if the data includes accented or non-Latin text.
  • Choose standard CSV when another system, script, database tool, or data pipeline expects a conventional CSV file and you do not want an Excel-oriented encoding marker.
  • Use an explicit import in Excel when data types, leading zeroes, dates, large numbers, or delimiters matter more than quick opening.
  • Use a different export format when you need formatting, formulas, multiple sheets, or a workbook that is ready to present.

If you are unsure how a downstream program handles a BOM, check its import documentation or try a small sample. Most modern text-processing tools support UTF-8, but a particular legacy parser may treat the BOM as part of the first column name or value if it does not recognize the marker.

Downloading Results Versus Unloading Data with SQL

There is an important distinction between downloading a result from Snowsight and exporting table data with Snowflake SQL. A Snowsight result download is a user-interface action with choices intended for convenient consumption. In SQL, a common pattern is COPY INTO a stage or external location with a CSV file format. That process is configured through Snowflake’s CSV file-format options; CSV for Excel is not a separate SQL data type that you select in place of TYPE = CSV.

SQL unloading is generally the better fit for repeatable or large-scale exports. It can be incorporated into a scheduled workflow and configured to control properties such as delimiters, quoting, headers, null representation, and compression. The correct settings depend on the receiving system. For example, a downstream loader may expect a header row, while another process may require a particular delimiter or handling for empty fields.

Do not assume that choosing a CSV file format in SQL automatically reproduces every detail of Snowsight’s CSV for Excel download. If Excel compatibility is a requirement for an automated export, verify the produced file’s encoding and how the target Excel workflow opens it. A controlled import procedure is often more reliable than relying on users to double-click files with different regional settings.

Common Problems That the Excel Option Does Not Fix

CSV for Excel is useful, but it is not a general data-cleanup or workbook-conversion feature. These issues often need separate attention:

  • Columns appear in one field: Excel may be using a different list separator from the comma in the file. Regional settings can affect how directly opened CSV files are split. Use the import preview and select the delimiter that matches the export.
  • Leading zeroes disappear: Excel may interpret a value such as 00127 as a number. Import the column as Text if those zeroes are meaningful.
  • Long numbers change: Excel has limits on numeric precision and may display long values in scientific notation. Treat identifiers as text rather than as quantities for calculation.
  • Dates change format: A date-looking string can be interpreted according to Excel’s locale or automatic type detection. Import it as text if the original character sequence must be preserved, or deliberately convert it using an agreed date format.
  • Characters still look wrong: Check the source data and the import encoding. A BOM cannot restore characters that were lost before export or fix a mismatch in a later system.

These behaviors are not unique to Snowflake. CSV stores text separated into fields; it does not carry the same typed-column and display-format information as an XLSX workbook. When exact interpretation matters, agree on a schema and import process with the person or system receiving the export.

Security and Safe Spreadsheet Use

CSV files can also carry a spreadsheet safety risk. If exported text begins with a character Excel treats as a formula indicator, such as an equals sign, Excel may interpret that cell as a formula when the file is opened. This is sometimes called CSV or formula injection. It matters most when a file contains untrusted user-supplied text that is later opened by someone else.

The Excel-specific download choice should not be treated as a security filter or as a promise that every cell will be imported as inert text. For exports containing untrusted content, consider whether the data needs to be neutralized before it is shared, and follow your organization’s data-handling rules. Any mitigation that changes a value can affect data fidelity, so choose it deliberately rather than applying an unexplained transformation. Also remember that a CSV is plain text: it does not protect sensitive data with workbook passwords or encryption.

CSV for Excel Compared with XLSX

If the goal is simply to move query results into cells, either CSV option can be convenient. If the goal is to deliver a polished spreadsheet, CSV is the wrong format. An XLSX workbook can preserve worksheet formatting, column widths, formulas, multiple sheets, and Excel features. A CSV contains a single plain-text table and does not preserve those workbook properties.

That distinction also affects how you share the data. A recipient can open a CSV in many applications, which is useful for interchange, but the recipient may see a different interpretation depending on software and locale. An XLSX file is more suitable when you want the recipient to see a prepared Excel workbook. If you download CSV and then save it as XLSX in Excel, review the imported values first: saving the file as a workbook does not undo incorrect conversions that happened during opening.

Frequently Asked Questions

Is CSV for Excel different from a normal CSV?

It is still a CSV-style text export. The Excel-labeled option is intended to improve Excel’s handling of UTF-8 text, commonly by adding a BOM; it is not an XLSX workbook and does not add spreadsheet formatting.

Does CSV for Excel change the Snowflake data?

It is intended to export the same query results. Its Excel-oriented encoding cue does not change the underlying table or add data types, though Excel can still reinterpret values when the file is opened.

Will CSV for Excel preserve leading zeroes?

No. A BOM helps with text encoding, not with Excel’s automatic conversion of number-like values. Import the relevant column as Text if zeroes at the beginning are significant.

Should I use CSV or CSV for Excel for an automated pipeline?

Use the format and encoding your receiving software expects, and test a representative file. For repeatable Snowflake exports, configure the SQL unload and the downstream import deliberately instead of relying on a person to open a download manually.

Can I turn a Snowflake CSV into an Excel workbook?

Yes. Open or import the CSV in Excel, check that the values and columns were interpreted correctly, then save it as an Excel workbook such as XLSX. The workbook can preserve spreadsheet features added in Excel, but the original CSV did not contain them.

Related

Try it: the Date Difference Calculator: days, months, years and weekdays between two dates, with the DATEDIF formulas.

More in How-to basics

Sources: Function behaviour and menu paths: Microsoft Support documentation for Excel. Method & sources

Spotted an error in this guide? Let us know.