3 min read
Categories Guides

Manual Translation vs CSV Import: Which Workflow Scales?

Manual editing and CSV import are not opposing philosophies. They are tools for different kinds of change.

An editor working on one important landing page benefits from seeing the final layout and choosing every phrase in context. A catalog manager changing specifications across 2,000 products needs a source that can be reviewed, filtered, and applied consistently.

The useful question is not “Which method is faster?” It is “Where does judgment happen, and how many records must follow the same rule?”

Manual editing works well for high-context content

Edit directly in WordPress or WPML when:

  • the number of records is small;
  • layout and surrounding page context shape the copy;
  • every item needs an individual editorial decision;
  • changes are rare;
  • the reviewer needs immediate front-end feedback.

Home pages, campaign pages, legal notices, and a handful of flagship products often belong here. A spreadsheet can hide the nuance that matters most.

The weakness appears when the same structured change must be repeated hundreds of times. It is difficult to prove that every product received the new specification, and correcting a systemic mistake means visiting the same screens again.

CSV or Sheets works well for repeated structure

Use a structured import when:

  • many products change together;
  • suppliers provide prices, stock, or specifications as files;
  • translators review catalog copy outside WordPress;
  • the team needs to filter missing values;
  • updates must match existing products through stable keys;
  • the operation will be repeated.

For WPML, Matrix Mode can keep translations in language-suffixed columns:

sku,title|en,title|pl,regular_price
DRILL-100,Power Drill,Wiertarka,149.99

This does not automate translation. It makes the batch visible: one row identifies the product, translated fields sit side by side, and shared values are not copied into separate files unnecessarily.

A hybrid workflow is often the most practical

A strong catalog process can use both approaches:

  1. Prepare structured facts and first-pass translations in CSV or Google Sheets.
  2. Review terminology and missing values in the table.
  3. Import a small staging sample.
  4. Inspect representative products in their real layout.
  5. Make individual editorial adjustments where presentation needs them.
  6. Keep the source of truth clear for the next batch.

The last step matters. If a translator changes WordPress but the Sheet remains authoritative, the next import may replace that edit. Decide which system owns which fields.

Scale depends on stable identity

A large spreadsheet is not scalable if the importer cannot tell whether a row is new or existing. For normal cross-store products, use SKU. Matrix ID / External ID can identify multilingual Matrix records. Avoid WordPress post IDs copied from another site.

Repeat the staging import. The second run should update the same products. That proof is more valuable than the speed of the first run.

Automation comes after the manual proof

Shirely 1.3 can save supported import and catalog/content export work as Automation slots. Run now and Run History make the saved configuration testable; WP-Cron or WP-CLI can execute it later.

Do not schedule a workflow merely because the button exists. First prove the source, mapping, identity, and result manually. WP-Cron is traffic-driven, and overlapping jobs that touch the same products, terms, or WPML/WCML groups should be sequenced.

Use a decision rule the team can remember

Choose manual editing when the value lies in individual judgment and page context. Choose CSV or Sheets when the value lies in consistent treatment of many structured records. Combine them when a batch needs both data discipline and editorial polish.

For a real Matrix example, see how to import products with WPML translations. If translators work in a shared table, review the Google Sheets translation workflow.