Count Real Shelf Stock and Update Shopify via CSV

Shopify only knows what you tell it. Every untracked event pushes the number in your admin away from what is physically on the shelf: a return that never got restocked, a box that arrived dinged and was quietly pulled aside, a picker who grabbed the wrong variant of two near-identical products, a receiving session where 24 units were entered as 42. None of these show up anywhere. With a few thousand SKUs, even a small monthly error rate means dozens of products where Shopify's availability is simply wrong - and you have no idea which ones.

When the drift runs in the wrong direction, you oversell. The storefront says 2 in stock, the shelf has 0, and an order comes in that you cannot fulfill. Now you write the apology email, process the refund, and hope the customer does not leave a one-star review that says "they sold me something they didn't have." For a small online retailer, those reviews are disproportionately expensive - they sit on your store forever and cost you the repeat buyers your margins depend on.

So you defend yourself, and the defense costs money too. You pad safety stock, listing fewer units than the system shows so an error cannot become an oversell - which means turning away sales on stock you actually have. Meanwhile the drift also runs the other direction: products listed as sold out with units sitting right there on the shelf. That is dead capital, invisible to your storefront and to you, quietly accumulating in every aisle.

The obvious fix - just count the shelves - does not survive contact with Shopify's admin. Counting with a clipboard and then updating products one at a time means searching for each product, opening the variant, clicking into inventory, typing the number, and saving. At 30-60 seconds per SKU, a 2,000-SKU catalog is a 20-hour data entry job, so it never happens. Shopify does support bulk inventory updates via CSV import, but filling that CSV by hand from paper count sheets just reintroduces the transcription errors you were trying to eliminate.

DataScan turns the physical count into a scan-and-type exercise. In Single Value Scan mode, you scan a product's barcode - the same EAN or UPC printed on the box that Shopify already stores in the product record - and type the quantity you see on the shelf. Scan, type, next. A hundred SKUs take 15-20 minutes. The app works completely offline, so a stock room with steel shelving and no signal is not a problem, and every count can be reviewed before export. One tap exports the whole session as a CSV or Excel file.

The Shopify side uses functionality built into every Shopify plan: inventory export and import via CSV. In your admin, go to Products, then Inventory, and export the inventory CSV for your location (choose the "All states" format - it is the one Shopify recommends). Open that file in Excel or Google Sheets next to your DataScan export, and use XLOOKUP or VLOOKUP to pull each SKU's counted quantity into the "On hand (new)" column. One important detail: fill in "On hand (new)", not "On hand (current)" - Shopify ignores the current column on import. Then import the file back into Shopify. Done. Storefront availability now matches the shelf, with no Shopify app installed, no API keys, and no integration to maintain.

Shopify's import has a built-in safety net that makes this workflow robust for a live store. When both "On hand (current)" and "On hand (new)" are present in the file, Shopify compares the current value in the file against the actual current value in your store - and rejects any row where stock changed between your export and your import, emailing you the details. So orders that come in while you are counting do not get silently overwritten; you just recount those few items and re-import. Since inventory imports cannot be undone, keep the original exported CSV as your backup before every import.

One honest limitation: DataScan is file-based. There is no live sync and no direct Shopify integration - the CSV merge is a manual step you do in a spreadsheet, and that is the trade-off that keeps the whole setup simple and cheap. If you want to remove even that step, DataScan can upload its exports to an SFTP server automatically, and third-party Shopify inventory apps (EZ Inventory, for example) can pull quantity updates from an SFTP server or URL on a schedule. That combination gives you a hands-off pipeline from shelf scan to storefront - but it is an optional advanced setup through a third-party app, not something DataScan does on its own.

Meeple Depot is an online-only board game retailer in Denver. Owner Tom stocks about 2,200 SKUs - games, expansions, card sleeves, dice sets, and hobby paints - in a 1,200 square foot warehouse unit, and sells exclusively through his Shopify store. It is a two-person operation: Tom plus a part-timer who packs orders in the afternoons. Stock drift came from everywhere: returns that sat in a corner instead of going back on the shelf, dinged boxes pulled aside to sell as damaged copies but never adjusted in Shopify, and pick errors between expansion boxes that look nearly identical on the shelf.

The cost was very visible. Meeple Depot was cancelling 5-7 oversold orders per month, each one a refund, an apology email, and occasionally a public one-star review that specifically mentioned the cancelled order. To protect himself, Tom started padding: fast-moving games were listed with one or two fewer units than the system showed, and anything questionable was marked sold out early. That stopped some oversells but meant turning away sales on games that were sitting right there on the shelf. The only real reconciliation was a full count once a year over a miserable weekend - and the numbers drifted visibly wrong again within weeks.

The new routine took one Monday to set up. Tom split the warehouse into six zones of roughly 350-400 SKUs each. Every Monday morning he counts one zone with DataScan in Single Value Scan mode: scan the EAN on the back of each game box - the same barcode already stored in his Shopify product records - and type the shelf count. A zone takes about 50 minutes. Then the desk work: export the DataScan CSV, export Shopify's inventory CSV in the "All states" format, XLOOKUP the counted quantities into the "On hand (new)" column by SKU, and import. The whole cycle runs about 70 minutes a week, and the entire catalog gets physically verified every six weeks.

The first full pass through all six zones was an education. Out of 2,200 SKUs, 163 were wrong in Shopify. Seventy-one phantom units - storefront said available, shelf said empty - each one an oversell waiting to happen. And more than 90 units of the opposite: real stock listed as sold out or below the true count, roughly $1,400 of retail value sitting invisible on the shelves. The import process also proved its safety net early: on one import, two rows came back rejected because customers had ordered those games between export and import. Shopify emailed the details, Tom recounted the two items, and re-imported - nothing got overwritten.

Six months in, oversold orders are down from 5-7 per month to roughly one per quarter - only ever from same-week drift in a zone counted five weeks ago. Tom removed all the manual padding and stopped marking items sold out early, so fast movers sell down to the last real unit. No new reviews mention cancelled orders. He has looked at the advanced option - DataScan exporting to SFTP with a third-party Shopify app pulling quantity updates automatically on a schedule - but for now, 70 minutes on a Monday morning is a price he is happy to pay for a storefront he finally trusts.