Size AI's Label Scanner reads the labels and tags on a garment and pulls 15+ structured data points, including brand, multi-format sizing, fabric composition, care symbols, and 5 stretch levels.
A jeans tag can print its size as W30 L32, as a letter size or as an EU number. Typed into a listing by hand, that is a chance to get the size wrong, and the size is already on the label. That gap is the whole reason the Label Scanner exists. Photograph each label and tag on the garment, tap the checkmark, and Size AI fills in the fields for you. The label always told the buyer that information. Now it tells the listing too.
What follows is what the Label Scanner pulls, how to scan a label cleanly, and a handful of edge cases worth knowing about.
What the scanner actually does
Your iPhone reads the text on each label. An AI model then sorts it into fields (brand, model, size and size system, fabric percentages, made in, season, care), and a separate on-device model recognizes the care symbols. The fields fill in the "Label Information" section of the Results screen and save with the garment.
One scan can fill 15+ data points across all the labels you photograph. The AI also corrects common misreads, like an O read as a 0 or a misspelled fiber name.
What gets extracted
Scanning a clean label produces:
| Field | What it reads |
|---|---|
| Brand and model | Full names plus model where available, like "Levi's 501" or "Patagonia Better Sweater" |
| Multi-format sizing | W30L32, Size L, EU 42, UK 10: the size and its system (US, EU, UK or JP) |
| Fabric composition | Each fiber and its exact percentage, like Cotton 84%, Polyester 15%, Spandex 1% |
| 5-level stretch analysis | Minimal, Light, Medium, High, or Maximum, derived from fabric content and construction |
| Product and style codes | Style numbers and item numbers for inventory tracking |
| Manufacturing details | Country of origin (Made in), season or collection, RN numbers |
| Care instructions and symbols | Wash temperature, drying method, ironing temperature, plus the pictograms most buyers don't bother to decode |
Care instructions matter more than they sound. They answer a buyer question before it gets asked. Put them in the listing description and you skip the message thread.
How to scan a label
The flow lives in the Results screen after a measurement capture.
- 1
Tap "Tag" or "Scan Label Now"
On the Results screen, tap "Tag" in the row under the photo, or "Scan Label Now" in the "Label Information" section. The label camera opens and focuses at close range.
- 2
Photograph every label and tag
Get close enough that the text is sharp, with the phone square to the label. In dim light, tap the flashlight button. Tap the shutter once for each label: the brand label, the size tag, and the fabric and care labels. The counter shows how many you have, up to 5.
- 3
Tap the checkmark
When every label is in, tap the checkmark. The camera closes and the "Label Information" section fills in a few moments later.
- 4
Check "Label Information"
Brand, size, stretch, fabric, care symbols and manufacturing details appear in the "Label Information" section. To correct or add a field, tap "Edit Label", make the change and tap "Save".
- 5
Use it in your listing
The label details feed AI Descriptions. To save them as an image in your Photos, tap "Download Image".
In the app's own estimate, the Label Scanner saves about 3 minutes per garment over typing brand, size, fabric and care by hand.
Tips that actually move the needle
- Flat label first. A folded or curled tag can hide text. Pull the label flat against the table or the inside of the collar before you take the photo.
- Every tag in one go. Labels photographed in the same session are read together and combined into one record. "Retake" starts over and replaces the label details, so photograph every tag again.
- Scan with a connection. Your iPhone reads the text, but sorting it into fields happens online. Offline, the app shows "Unable to analyze label. Please check your connection and try again."
- No label? Cut or missing tags can't be scanned, and label fields are edited after a scan. AI Descriptions still works from the photo and measurements.
Note: Any language on the tag
The OCR reads labels in any language and returns the extracted fields in English, so a Japanese care label and a French fabric tag land in the same structured record.
