Scan-to-Autofill: Adding New Products From One Photo With AI

AI product enrichment from a photo is a feature that reads one picture of an item and suggests catalog details — a product name, a description, and a category — instead of a staff member typing each field by hand. In Retailer OS this is called scan-to-autofill, and it ships as part of a Messaging & AI plan (Starter $39.99/month, Growth $69.99/month, or Pro $99.99/month). It is not included in the base store price, and it is not a separate add-on you buy on its own — it comes with whichever Messaging & AI tier you choose.

What Is AI Product Enrichment From a Photo?

Product enrichment is the process of turning a raw image of an item into structured catalog data — the name, description, category, and other fields a point-of-sale and online store need before the item can be sold. Without it, someone has to look at the product, decide what to call it, write a description, and pick a category every single time new stock arrives. With it, the photo does the first draft and a person reviews it before it goes live.

This matters most for stores that add inventory constantly and in small batches: boutiques receiving a new vendor drop every week, consignment and resale shops where every item is unique, or any shop where one person is responsible for getting new stock onto shelves and into the online store at the same time. Typing out 40 new SKUs by hand is slow and it's where typos and inconsistent naming creep into a catalog.

How Does Scan-to-Autofill Actually Work?

The workflow sits inside the normal process of adding a new item to Retailer OS. Instead of opening a blank item form and filling in every field, a staff member takes a photo of the product during intake or receiving. The AI reads the image and returns suggested values for the item's name, description, and category. Those suggestions land in the same item form everyone already uses — nothing is auto-published without a look first.

From there, the person adding the item reviews what the AI suggested, edits anything that's off, fills in the fields that need a human decision — price, SKU, barcode, quantity on hand, and which variants apply — and saves it. Once saved, the item is a normal row in the inventory ledger: it has per-location stock like every other item, and it's ready to sell at the register or publish to the online store if you run one.

What Product Details Can AI Fill In From One Photo?

Scan-to-autofill is aimed at the fields that take the most time to write from scratch and that a photo can reasonably inform:

• Product name — a usable working title for the item based on what's visible in the photo
• Description — a starting draft of selling copy, which you can edit before it's used anywhere customer-facing
• Category or type — a suggested place for the item in your catalog structure, so similar products group together consistently

These are suggestions, not final answers. The point is to replace a blank form with a reasonable first draft, not to remove the review step. If you want AI help specifically with sales copy once an item is already in the catalog, that's a related but separate workflow covered in AI product descriptions for retail.

Where Does This Save the Most Time?

Scan-to-autofill pays off most in situations where products arrive one at a time or in small, varied batches rather than as a clean spreadsheet from a vendor:

• Consignment and resale — every item is different, there's no vendor catalog to import, and someone has to describe each piece from scratch
• New seasonal buys — a boutique or gift shop bringing in a vendor's new line one box at a time, before a formal catalog file exists
• Walk-in trade-ins or buybacks — books, instruments, sporting goods, or jewelry coming in from customers rather than suppliers
• Multi-location catalogs — keeping naming and categorization consistent when several staff members across stores are each adding items independently

If your new stock does arrive as a clean vendor file or spreadsheet, CSV import is usually faster than photographing each item one by one — see the fastest way to add products to inventory for when to use which method.

What Still Needs a Human Check?

AI product enrichment drafts text; it does not make merchandising or pricing decisions, and it doesn't replace the judgment calls that keep a catalog accurate:

• Price — the AI doesn't set a price; that's still a decision you make based on cost, margin, and market
• SKU and barcode assignment — scanning or generating a barcode is a separate step; see how to set up SKUs and barcodes for the basics, and GS1 if you need to register for globally unique barcodes
• Quantity on hand — a photo tells you what the item is, not how many you have; counts are entered separately
• Anything a photo can't show — materials, serial numbers, age restrictions, or compliance details for regulated categories still need a person to confirm

Treat the AI's output the way you'd treat a draft from a new employee: usually a good starting point, always worth a second look before it goes live.

How Much Does AI Product Enrichment Cost in Retailer OS?

Scan-to-autofill is one of the features included on every tier of the Messaging & AI plan: Starter at $39.99/month, Growth at $69.99/month, or Pro at $99.99/month. Every tier includes every AI feature — the product enrichment tool, the AI merchandising assistant, demand forecasting, and reorder suggestions — along with texting, email marketing, and the shared inbox. The tiers differ only in how much messaging and AI usage is included each month, not in which features you can access. There is no way to buy scan-to-autofill by itself outside of a Messaging & AI plan, and it is not part of the base store price.

If you're weighing whether the AI tools are worth the add-on cost across your whole operation, AI for retail: practical use cases that actually save time walks through the other features on the same plan, including AI demand forecasting for reordering.

How Do I Turn On Scan-to-Autofill for My Store?

• 1. Subscribe to a Messaging & AI plan from Setup — any tier unlocks the full set of AI features, including scan-to-autofill.
• 2. Take a photo of the new item during intake or receiving, using the item form's photo capture.
• 3. Review the AI's suggestions for name, description, and category, and edit anything that needs correcting.
• 4. Fill in the fields AI doesn't set — price, SKU, barcode, variants, and starting quantity — then save.
• 5. Print a barcode label for the shelf and the item is live across every register and location that shares the catalog.

Because the item lands in the same inventory ledger every other product uses, there's no second catalog to maintain — it's on-hand per location the moment it's saved, and it's ready to publish to an online store if you run one.