Duplicate products in Shopify — find them, merge the stock, stop the leaks

Duplicate products in Shopify are two or more product records for the same real-world item — usually created by a re-import, a staff re-creation, or a supplier catalog sync — and each duplicate quietly splits your data in two: stock divided across records, sales history fragmented, and velocity per record understated. A SKU showing 6 units on record A and 4 on record B reads as "10 total, fine" until record A sells out and its customers see nothing available. Here's the click-path to find and merge them, and the rules that stop them growing back.

The click-path: find the duplicates

Walk 1 — SKU sort (five minutes, catches most cases):

1. In Shopify admin, go to Products.
2. Click the SKU column header to sort ascending; blank-SKU products cluster at the top (as of October 2026 the products list is sortable by SKU column — the Help Center documents sorting and filtering the product list by column — Shopify Help Center, "Sorting and filtering products," accessed October 2026; if your view hides it, click the columns/eye icon to add SKU).
3. Scan for identical or near-identical SKUs (CASE-IP15-BLK and CASE-IP15-BLK-1, or the same code with a trailing space).
4. Sort descending and repeat — duplicates created by import errors often share an exact SKU, which Shopify will have accepted because its SKU field is not natively unique-enforced (Shopify Help Center, "SKUs," accessed October 2026).

Walk 2 — the export check (thorough, for bigger catalogs):

1. Products → Export → CSV (mechanics of the export loop, columns and re-import gotchas, are owned by the bulk edit and CSV guide — this walk only uses the file).
2. In the spreadsheet, find the Variant SKU column (insert its actual letter — it varies by export version), then build a helper column: =TRIM(UPPER(S2)) over that column to normalize whitespace and case (the formulas below assume SKU is in column S — substitute your letter).
3. Add a count helper: =COUNTIF($S$2:$S$5000,S2) and filter for values >1. Every row with a count above 1 shares a SKU with another variant.
4. Also scan the title column for stem-duplicates: sort by product title and eyeball the blocks where the same name repeats with different handles (/case-ip15-black vs /ip15-black-case-2).

Walk 3 — the demand split check (finds duplicates even without matching SKUs):

Sort sales-by-product reports for your best categories and look for two products with suspiciously alternating sales — one selling weekdays, one weekends, neither with a full story. That pattern is often one real product split in two.

The click-path: merge without losing stock or history

Shopify has no one-click "merge products" as of October 2026 (the Help Center's duplicate/merge guidance has no built-in merge; Shopify Help Center, "Editing product variants" / merge guidance, accessed October 2026), so a merge is: consolidate stock onto one record, redirect the loser, and preserve history by hand. Pick a survivor — the older record with more sales history.

1. Products → open the survivor. Confirm its SKU, cost per item, and barcode are the correct, canonical values (one SKU per real-world item — the identifier logic is in what is a SKU number).
2. Open the duplicate. Note its on-hand quantity per location, then set its on-hand to 0 and record the move.
3. On the survivor, add the duplicate's units via Inventory → Adjust (or Edit quantities), entering the transferred count. Now the physical truth lives on one record.
4. On the duplicate: if it has sales history you want preserved, keep the record but set status → Archived and rename the title to prefix DO NOT USE — merged into [survivor SKU]. Archiving removes it from storefront and order flows while keeping its history queryable — the Help Center documents that archived products are hidden from all sales channels but remain in admin with their records intact (Shopify Help Center, "Archiving or deleting a product," accessed October 2026). If it has no meaningful history, delete it outright.
5. Check for references: open the survivor's page and confirm images, metafields, and collection membership are complete — duplicates often mean reviews, images, or tags got scattered across both records.
6. Run one order test: place a draft order against the survivor to confirm quantity, price, and cost per item all read correctly.

If the duplicate has variants (e.g., the same phone case exists as a 3-variant product twice), consolidate by adding missing variants to the survivor first, moving quantities, then archiving — variant-level merges follow the same pattern one variant at a time.

Where duplicate stock actually hurts

• Oversell exposure halves your real availability math. Two records each showing 4 units when only 4 exist physically (2 + 2 split) is the classic duplicate inventory state — the total is real, the per-record numbers are fiction, and each record's reorder point is blind to the other's sales.
• Velocity per record drops below truth, which suppresses reorder triggers on exactly the SKUs that sell well enough to be worth duplicating.
• Counts produce phantom variances. A cycle count finds 8 physical units, the summed records show 8, but each record individually disagrees with its bin — producing discrepancies that dissolve only when you notice the second record.

Prevention rules

Duplicates regrow from specific seeds — cut the seeds:

1. Never re-import a catalog in full to "update" it. Updates go through CSV re-import matching on existing handles/SKUs (the bulk-edit guide owns that procedure), never through a fresh product creation import.
2. One staff rule: search before creating. Before adding any product, search the admin for the title and SKU. Most duplicates are created in ten seconds by someone who didn't search first.
3. Supplier catalog syncs get reviewed, not trusted. Any app or feed that creates products should be checked monthly with Walk 2 — sync tools are the leading industrial-scale duplicate factory.
4. SKU conventions prevent accidental twins. If CASE-IP15-BLK and CASE-IP15-BLK-1 can both exist, the naming scheme has a hole — the structural fix is in SKU naming for Shopify.
5. Quarterly duplicate sweep. Walk 2 takes ten minutes at 500 SKUs and catches everything the month-by-month drift produced.

Common questions

How do I find duplicate products in Shopify? Sort the products list by SKU and scan for repeats or blanks, or export the product CSV and use a COUNTIF over normalized SKUs to flag every duplicate pair. Also sort by title for stem-duplicates with different handles, and check sales reports for split-demand patterns.

Can you merge two products in Shopify? Not with one click as of October 2026. The manual merge is: choose a survivor, move the duplicate's on-hand quantity onto it, archive the duplicate (preserving sales history) or delete it if historyless, and confirm images, tags, and collection membership live on the survivor.

Why are duplicate products a problem for inventory? They split one real-world item's stock across two records, so each record's on-hand and reorder point is blind to the other's sales — velocity per record understates demand, reorder triggers misfire, and cycle counts produce variances that only make sense once the second record is found.

What causes duplicate products in Shopify? Full catalog re-imports that create instead of update, staff re-creating products without searching first, and supplier catalog or sync apps generating new records on feed changes. The prevention rules — matching imports, search-before-create, monthly sync audits — target those three seeds directly.

Do duplicate SKUs cause overselling? They can: two records each showing part of the same physical stock lets both sell against their own number while the combined availability is wrong. Correcting the split consolidates the real quantity on one record and restores honest availability everywhere it's displayed.

Your next step

Run Walk 2 on your full catalog this week — the COUNTIF over normalized SKUs takes ten minutes and gives you the complete duplicate list, not just the ones you've noticed. Merge any pairs with the survivor protocol above, then set a quarterly sweep reminder and adopt search-before-create as the standing staff rule so the list stays empty.