Frequently asked questions

Not much. Download the template for the type you want and it arrives with a correct header row plus worked example rows, each demonstrating a different situation. Change the values, keep the headers, import it. The quick start walks through a real import in about five minutes.

If “definition” versus “value” is new to you, two paragraphs here will save you a lot of confusion.

Five things: metafield definitions, metafield values, metaobject definitions, metaobject entries and blog articles. Compared side by side →

Not products, collections, customers, orders, themes or files. Products have Shopify's own CSV importer; ImportKit covers the custom data layer around them.

Upload CSVs. Excel, Sheets, Numbers and LibreOffice all export one in a couple of clicks — here's the route through each, including the encoding setting that matters.

No. Mapping is explicit — you pick the Shopify field for each column from a dropdown. Any export from anywhere works without renaming a thing, and extra columns are harmless. More →

Install on a Shopify development store and run the real files there. Credit purchases on development stores are test charges, so a full rehearsal costs nothing. Failing that, import a five-row sample into production and check it in the admin before running the rest.

No. ImportKit never deletes anything, which also means it can't reverse itself. Undoing an import is manual work in the Shopify admin.

That's a deliberate design decision rather than a missing feature. An app with delete access that can bulk-remove data from a CSV is a much more dangerous thing to install than one that can only add and update. The trade we made instead is the free pre-import check — read every row's verdict before anything is written, as many times as you like.

For definitions, metaobject entries and metafield values, ImportKit recognises what already exists and applies your conflict mode: skip it, update it, or report it as an error. Skipped rows aren't charged, so a re-run of an already-imported file costs nothing.

Blog articles are the exception. Articles aren't uniquely keyed, so a second run creates them again. Only run an article file once.

Only if you choose Update mode. The default is Skip, which leaves existing data completely alone.

Even in Update mode, some things are never rewritten: a metafield definition's type, and a metaobject definition's existing fields. Those are immutable in Shopify and rewriting them from a spreadsheet would be destructive in a way that can't be undone.

The check catches most of it for free — a name in the key column fails key validation, a mis-mapped owner type fails on every row. Going back a step costs nothing, and nothing is written until you press Start Import.

No. Blank cells are skipped rather than written, so a blank can't clear a value. That's a safety choice: it means an export with gaps can't quietly wipe data. To empty a metafield, delete it in the Shopify admin.

Free to install. One credit per row written, with credits sold in one-time packs from 50 for $1.99. Pricing → · How credits work →

No free credits, but everything up to the moment of writing is free and unmetered — templates, uploads, mapping, and unlimited pre-import checks. So you can validate an entire file before spending anything. First pack is $1.99.

No. They stay on your store's balance indefinitely, and survive an uninstall-and-reinstall on the same store.

No. Rows that error, skip, conflict or get interrupted are refunded automatically. A credit is reserved just before each row is sent and released the moment the outcome is known, so an import that fails entirely costs nothing.

Run the pre-import check. It reports “N of M rows will be written”, and N is the credit cost. It's free and re-runnable after every edit.

The import stops cleanly at your balance and says so. Nothing was written that couldn't be paid for. Buy a pack, then press Retry — it resumes from the first unprocessed row and never re-charges rows that already succeeded.

Per store. Credits belong to the store that bought them and aren't transferable, so agencies buy credits on each client store. The charge lands on that store's Shopify invoice.

Up to 50 GB and 1,000,000 rows per import, and we can raise that on request. The pre-import check reads the whole file at once, so it caps at 50,000 rows or 32 MB — above that the check is skipped but the import still works, with conflict handling applied per row exactly as usual. All limits →

Rows go to Shopify one at a time, so expect thousands of rows to take minutes rather than seconds. Serial processing is deliberate — Shopify's API throttles hard enough that parallel writes end up slower, and one-at-a-time keeps every row's outcome unambiguous.

No. Imports run as background jobs on our servers. Close the tab and check progress later under Jobs.

Yes. It stops between rows, keeps everything already imported, and charges only for those rows. Retry resumes from where it stopped. Cancellation is checked every few rows rather than instantly.

No. Every import is started by you, deliberately. There's no scheduler and no automatic sync — which also means nothing writes to your store while you aren't looking.

Shopify gates metafield definitions by the owner resource's own permission, not by a metafield permission. To let you define a metafield on a market, the app needs market write access — even though it never reads or changes a market. That's why permissions for discounts, locations and checkout customisations appear on a CSV importer. The full list, with reasons →

No, and that's on purpose. ImportKit doesn't request customer or order permissions, which keeps it out of Shopify's protected customer data review and means it cannot read that data at all.

Rows targeting those owner types aren't silently dropped — they're reported individually with the exact missing permission, and the rest of the file still imports.

The file itself is never written to disk — it's read as a stream as it arrives. The values from the columns you mapped are stored alongside the job record, so you can review per-row results and resume an interrupted import. Columns you ignored or left unmapped are dropped before that and never stored. Full detail →

Your session and access token are deleted immediately, so all API access ends. Import history and credits are kept at that point in case you reinstall, and everything is deleted when Shopify sends a shop redaction request — or sooner if you ask. Retention →

No analytics, no advertising, no session recording, no third-party trackers, and no error-reporting service. The only cookies are the ones Shopify's embedded app authentication requires.