AI tagging
Tag the products in every frame, automatically.
A two-pass vision pipeline detects candidate products and verifies them against your SKUs. Above 80% confidence auto-accepts; the rest queue for a human.
No code. Free plan, no card.
FRAME · 0:08UGC reel · @priya01
Detect, then recognise
Pass one finds candidate products and bounding boxes in the frame. Pass two verifies each candidate against your synced catalogue, so a match is a real SKU, not a guess.
Shoppable UGC
A pin becomes an order
WROGN Men Silver-Toned Watch
4.8 · 213 reviews
$24.76
Orders ledger
#4821$24.76attributed to@darcy.s
02
A threshold you control
Matches above 80% confidence are accepted automatically. Anything lower waits in the review queue with the suggestion attached, so a human confirms in one click rather than tagging from scratch.
Content
Everything collected, moderated and tagged before it reaches a widget
- ReelWROGN Men Silver-Toned Watch@sky.bApproved
- VideoAirlift Overcoat Brown@ines.wearsPending
- PhotoJute Tote Bag@marcus.kApproved
- ShortTennis Cardigan Spring/Summer@lena.jRequested
- ReviewBodycon Sunscreen SPF 50Google ReviewsNot needed
- PhotoHeldWhite Sweater Green Stripes@ayo.runsHeld
03
Cross-channel SKU mapping
A product mentioned on Instagram, TikTok, YouTube or a review maps to the same Shopify, WooCommerce or BigCommerce SKU, so the tag resolves to a real product page and a real price.
Sources
Nine sources, one feed, one script
- TikTok
- YouTube
- X
- Threads
- Google Reviews
Gallery · homepage
<script src="…/idukki.js"> · 37 KB- TikTok
- YouTube
- X
- Threads
04
Hotspots timed to the video
On video, tags carry a timestamp: the product appears in the drawer at the second it appears in frame, and the builder shows the hotspots exactly as shoppers will tap them.
Widget builder
Pick sources and a layout; the preview is the real renderer
Sources
- TikTok
- YouTube
- Hashtag
- Reviews
Layout
Products
AI tagging on
Add to cart inline
Live preview
How it works, from switched on to paying for itself.
- Step 1
Ingest the asset
From Instagram, TikTok, YouTube, the DAM, a hashtag campaign or a creator upload. Same pipeline for image and video.
- Step 2
Pass 1: detect
A fast vision model finds candidate products and bounding boxes inside the frame.
- Step 3
Pass 2: recognise
Each candidate is verified against your SKU set.
- Step 4
Accept and ship
Above the threshold, the tag is live. Below it, the suggestion waits for review in the queue.
“Being able to see how the clothes move and look on different women makes a huge difference.”
Three things to know before you turn the threshold down.
Anything below the threshold waits in the review queue inside the DAM, with the candidate product already attached.
- 01
Two-pass, not one-pass
Detect-then-recognise is slower than a single pass but catches multiple products per frame and aligns each confidence score with your own SKU set.
- 02
The confidence threshold is yours
Default 80% auto-accept. Raise it to 90% for high-AOV catalogues; lower it for catalogue exploration. It is a live setting, no re-run needed.
- 03
Re-tag when the catalogue changes
The tagger runs on every asset in your library continuously. Change SKUs and the hotspot timeline updates in place, with no re-import and no re-publish.
The questions to ask before you switch it on.
Which catalogues can it map to?
Any synced catalogue: Shopify, WooCommerce, BigCommerce, Wix, or a CSV/API feed for custom stores.
What happens below 80% confidence?
The suggestion goes to the review queue with the candidate product attached. A reviewer confirms or corrects it; nothing publishes untagged by default.
Does it work on video?
Yes. Tags on video carry a timestamp so the hotspot and the product drawer appear at the right second.
Is AI tagging on every plan?
Product tagging on image and video, multiple product tagging and hotspot tagging are listed for every band, including Free, in the comparison table at /pricing. The AI Tagging API is also available standalone on the API platform.