Numbers 2026: How to Fix Chinese Text Garbling in Cross-Border Product CSVs?

The product names look like unreadable characters in Numbers, or several fields have shifted into the wrong columns.

Fast fix: Keep the original CSV unchanged, import a working copy, check text encoding and separators separately, then verify the exported copy before sending it back.

Who this is for: Cross-border sellers repairing multilingual product data without risking the original catalog.
Shopify operators checking a product CSV on a Mac before an import or update.
Reviewers deciding whether the problem is display, encoding, or file structure.

SECTION 01 Numbers CSV Chinese text garbling 2026: identify the failure before editing

Treat garbled text as a diagnostic problem first, not as a request to retype product data. A CSV can look wrong because Numbers interpreted its text encoding incorrectly, because the row and column structure was parsed incorrectly, or because the source already contains damaged characters. These causes need different fixes.

Use this timeline to control the risk:

• Before editing: Preserve the original and create a separate working copy.
• During diagnosis: Compare a small set of fields and rows; change one import setting at a time.
• Before return: Export another copy and verify its encoding, columns, and product relationships.

Numbers’ import settings let you adjust text encoding and separators for delimited text. That means a bad first display does not, by itself, prove that the CSV is damaged. Apple documents these controls in its Numbers guide to importing text files.

Start with representative values rather than the whole catalog. Choose a product name containing Chinese text, a description with punctuation, an option or variant value, and—if present—a row with an image URL. Compare those values against the source system or a known-good export. Don’t use a garbled Numbers display as the source for reconstruction.

First, preserve the evidence

1. Keep the downloaded or received CSV unchanged. Make a working copy in a separate folder.
2. Give the copy a clear filename so you can distinguish it from the original and from any later export.
3. Open the copy in a plain-text viewer or another reliable CSV reader. Check the same sample values you selected for comparison.
4. Record what you see: whether the text is wrong everywhere or only in Numbers, whether headers are recognizable, and whether values appear under the expected columns.
5. Do not save changes over the original while testing import settings.

A second viewer is a comparison, not an automatic verdict: it may also guess the encoding. If the two displays disagree, test a fresh copy in Numbers with a different text-encoding selection. If both show the same incorrect content, compare the CSV with the original product source before assuming an import preference will fix it.

SECTION 02 What does the symptom tell you?

Use the visible failure to choose what to inspect next. Don’t change encoding, separators, and the product data all at once; doing so makes it harder to tell which change helped and can hide a structural problem.

These are diagnostic paths, not a claim that one setting always fixes a given symptom. A file may have both an encoding problem and a separator problem. Shopify’s product CSV format guidance sets out the platform’s UTF-8 requirement and file-structure expectations; verify the exported file against that guidance rather than relying on how the worksheet looks.

SECTION 03 How do you repair the working copy without changing product data?

Reimport the copy and test encoding

In Numbers, open the working copy as a text file and review the import settings before completing the import. Find the text-encoding control in the dialog. If the current selection produces garbled values, choose another plausible option and preview the same sample fields again.

Don’t assume a single encoding is correct for every supplier, export tool, or language mix. Use the version that makes the values match the source of truth, not simply the one that makes the grid look more familiar. If the dialog does not show the controls you expect, consult Apple’s text-file import instructions; interface details can change between software versions.

Once the preview matches, import into the working copy. Check the original product names, descriptions, option values, and headers again inside Numbers. If a field cannot be verified against a trusted source, leave it untouched and ask for a clean export instead of guessing at the intended characters.

Separate encoding checks from column-structure checks

If the text is readable but everything sits in one column, or values appear under the wrong headings, focus on separators and text qualifiers. A comma can mark a new field, but a comma inside a product description may be part of the value. Quoted fields and line breaks inside quoted content also affect how a CSV reader recognizes records.

RFC 4180’s CSV format description specifies, among other structural details, how fields containing commas, quotation marks, or line breaks can be enclosed and how embedded quotation marks are represented. In Numbers, review the separator and text-qualifier controls independently from the encoding control. Compare the header with a product row and a variant row before editing further.

A practical test is to inspect a small, representative section:

• Does each header occupy its own expected column?
• Do product identifiers and their associated values remain in the expected fields?
• Do descriptions containing punctuation stay within one field?
• Does a row containing a variant remain associated with the correct product?
• Do image URLs still appear in the intended column?

If only one row breaks the pattern, don’t “repair” the entire sheet by shifting columns manually. Identify whether the original CSV contains an unexpected quote, delimiter, or line break. Shopify also documents common product CSV import issues; compare the observed error with its current guidance before making structural edits.

Keep a clean export as a separate milestone

When the working copy displays and aligns correctly, save or export a new file. Do not overwrite the source CSV. If you have to make additional corrections, keep those in the working copy and create a new output copy after review.

UTF-8 is the required encoding for Shopify product CSV files, according to Shopify’s CSV format documentation. UTF-8 files may also include a byte order mark; Unicode explains the role and behavior of the UTF-8 BOM. Don’t add or remove a BOM as a guess. Check what the export produced and follow the platform’s current import requirements.

At this milestone, separate two acceptance checks:

1. Display check: The intended text and columns look correct in Numbers.
2. File check: The exported file has the expected encoding, headings, separators, and product relationships.

Passing the display check does not prove that the exported file meets the platform’s CSV requirements. Likewise, an import error does not necessarily mean the original product content was wrong. Treat these as separate checks and retain the original until the exported copy passes review.

SECTION 04 Before you return the Shopify product CSV

Review the exported file, not just the open Numbers document. The minimum review should cover the column headings, a sample of multilingual values, product identifiers, variant associations, and image URLs. Compare those fields with the original export or another authoritative source. Shopify’s format and import guidance can change, so check its current product CSV instructions before a production update.

Be especially cautious with sorting. A row sort that ignores the full table or a selection that excludes related rows can break the relationship between product data, variants, and images. Before returning the file, compare those relationships with the original. If an identifier, variant association, or image URL has moved to a different product, stop. Restore a clean copy and repeat the edit; don’t upload a file whose relationships you cannot verify.

For a low-risk review, use a limited test file or the platform’s available preview and validation steps before a broad update. Confirm that the columns are accepted and that the displayed products match the expected records. Shopify’s common import issue guide is useful when the file looks correct locally but the import reports a format or data problem.

If your team regularly repeats this process, document which source produced the file, which import settings were used, and who checked the exported copy. That record helps the next reviewer distinguish a recurring source-file issue from a Numbers import choice. It also makes handoffs safer when one person repairs the file and another approves the update.

SECTION 05 FAQ

Why does a CSV look garbled in Numbers but normal elsewhere?

Numbers may have interpreted the text with a different encoding from the one used to create the file. Compare the same values in the original source and another viewer, then test a fresh working copy with a different encoding selected in Numbers. If the source itself contains incorrect characters, an import-setting change cannot reconstruct the original text.

Where can you change text encoding when importing a CSV into Numbers?

Open the CSV as a text file and inspect the import settings before completing the import. Apple’s Numbers documentation describes controls for text encoding and separators. Their exact labels and layout may vary by software version, so check the dialog shown on your Mac instead of relying on a menu path from an older tutorial.

How can you tell whether the file is damaged or Numbers has misread its encoding?

Compare a sample field against the original product source and inspect the working copy in another text viewer. If the source is correct but Numbers displays a different value, test another encoding during a fresh import. If incorrect characters already appear in the source or across independent viewers, stop editing and request a clean export before proceeding.

What should you check before returning a Shopify product CSV edited in Numbers?

Export to a separate file, then verify UTF-8, expected headers, separators, and the connection between products, variants, and image URLs. Check Shopify’s current CSV requirements and validate a small test before using the file for a broad update. Keep the untouched original available, and stop if any product relationship has changed or cannot be verified.

SECTION 06 Choose the next step based on how often you repeat the repair

If you only need to repair one file, a local Mac and a carefully preserved working copy may be enough. The trade-off is that the process depends on the available machine, its Numbers settings, and your team’s handoff method; inconsistent settings or shared-file permissions can make repeat reviews harder to reproduce. For an ongoing workflow, a remote Mac can give reviewers a shared macOS environment without requiring each person to use the same physical machine.

If you need to reproduce Numbers imports and product-file reviews as a team, compare that workflow with the VPSNIX remote Mac service and review the available rental plans before deciding. Renting is not the right choice for every seller: a team with a long-running, heavy workload or a need for local physical connections may be better served by a dedicated Mac. For temporary macOS access or repeatable review work, assess whether a remote environment fits your file permissions, review responsibilities, and return process.

SECTION 07 FAQ