You reduce ecommerce returns by asking the fit question before checkout instead of after. A conversational agent asks what someone usually wears, in their own words, types the answer into a structured fit record, and posts it to your stack, so the mismatch that would have shipped and come back gets caught first.
01 · They talk
02 · It becomes a record
typed · validated · deduped
03 · Your stack acts
Who starts it
Someone staring at a size chart with no confidence, or someone holding a box that is not what they ordered. Almost always mid-checkout or mid-unbox, rarely during a quiet moment.
What it costs today
A static size chart nobody trusts, so shoppers order two sizes and return one. A returns portal with six reason-code dropdowns and no room for "it fits like a size down". A wholesale inbox where one real buyer sits next to a hundred spam pitches.
What it has to produce
A typed record: the item, the fit or the fault, what the customer actually wants done about it, and whether anything about it needs a person right now rather than a form.
70%
of fashion returns come down to size or fit, and a static chart never asks the question that would have caught it
FitEz
| Field | Type | Required | Why it is there |
|---|---|---|---|
| item_of_interest | text | Yes | Anchors the fit note to the right product |
| usual_size_other_brands | text | Yes | Calibrates the recommendation to how they actually fit, not a chart |
| fit_preference | text | Yes | Relaxed, true-to-size or fitted, in their own words |
| order_number | text | Yes | Ties every return, exchange or report back to the order |
| return_reason | text | Yes | Routes the return and flags a defect pattern early |
| desired_size_or_color | text | Yes | What the swap should actually deliver |
| safety_or_allergen_concern | boolean | Yes | The one answer that triggers the hard stop to a human |
| business_type | text | Yes | Sorts a wholesale lead before anyone quotes a price |
| product_categories_interested | list | Yes | What the wholesale team should actually pitch |
| would_reorder | boolean | Yes | The honest signal a review widget never asks for |
And what it will not do
Collected data
212 collected34 this week5 need attention
Export CSVJordan Bell
Captured 40m ago · Find My Size · joina.chat
Captured data 5 of 6 fields
From the conversation
“I’ve ordered from a few different denim brands and sizes never seem consistent between them , so I never really know what to order until it’s too late to send it back easily.”
Every value keeps the sentence it came from, and a value the visitor never said is never stored. A Gnosari cannot quote itself into your data.
From here a record can go straight on to Slack, a sheet, a Zap or your own API. The inbox is where you read them, not where they get stuck.
Signed headers
Body
{
"type": "data.collected",
"event_id": "1b6f9c02-4d1a-4f7e-9b3c-2a5e8d0f1c44",
"created_at": "2026-08-11T09:14:22Z",
"payload_version": "v1",
"data_kind": "Fit note",
"record_id": 1234,
"attributes": {
"item_of_interest": "Canvas field jacket",
"usual_size_other_brands": "Medium in most brands, Large in slim cuts",
"fit_preference": "True to size, prefers not to size up",
"contact_email": "jordan.b@example.com"
}
}Two events
data.collected fires the first time a Gnosari collects a new kind of data. data.updated fires when a customer corrects something.
Signed, so you can trust it
Standard Webhooks HMAC-SHA256, with webhook-id, webhook-timestamp and webhook-signature on every request. Verify before it touches an order.
Quotes are opt-in
The provenance block, the customer’s verbatim words, is absent unless you turn it on for that destination.
Zapier or Make
Paste a Catch Hook URL into your Gnosari destination, then point the Zap at your store.
Free tier, no app to install, and it fires the moment the record is complete.
Straight to the team
Anything that accepts an incoming webhook takes the payload as it is.
No middleman. Support sees the record before the ticket does.
Your own endpoint
Give us an https address, verify the signature, act on it yourself.
A few lines of code, and the delivery log tells you when something did not land.
And you can see every attempt
| When | Event | Outcome | Detail |
|---|---|---|---|
| 19:04:22 | data.collected | Delivered | 200 OK |
| 19:04:18 | data.collected | Failed | 503, retrying in 30s |
| Yesterday 09:37 | data.updated | Delivered | 200 OK |
| Yesterday 07:52 | data.collected | Delivered | 200 OK |
Every attempt is kept, newest first, with the reason when there is one. Failures retry on their own, and the signing secret lives in your dashboard so you can rotate it without calling anybody.
These are examples, not a partner list. Shopify, BigCommerce, Klaviyo and Zendesk are all reachable the same way: through a Zap, or straight to an endpoint you control. If it takes an HTTP request, it takes the record.
A button in the corner that opens into a conversation. Your logo, your colours, sitting on the pages you already have.
No chat window and no widget. Your headline leads, the thread sits under it, and it reads like you asked rather than like software collecting.
Every screen on this page is a published Gnosari, embedded from the app. Type into any of them.
display_mode="bubble" A floating button in the corner. Opens over your page and closes again. The default, and the one nobody has to be taught.
Four corners to choose from
display_mode="sidebar" A fixed panel down one side, always open. For pages where the conversation is the point rather than an aside.
Left or right, any width
display_mode="sidebar-push" The same panel, but your content moves over to make room instead of being covered. Nothing on the page is ever hidden.
Left or right, any width
display_mode="drawer" Slides in over the page when it is called, then slides away. The middle ground between a bubble and a sidebar.
Left or right
01 · A link you send
Put it in an email, a text, a booking confirmation or a signature. Each link carries its own opening line and headline, and counts its own visits, so you can tell which channel actually worked.
02 · A code they scan
Every link comes with a QR. On a table card, a receipt, a delivery note, a door. Someone points a phone at it and the conversation starts, with no app and no typing.
03 · A flyer you hand over
A builder for the printed thing itself. Pick a template, restyle it, edit the words on the page and export a print-ready poster or card with the QR already placed.
Twelve curated looks, each a colour and a background pattern chosen to go together. Pick one and you are done.
22 colour themes and 44 background patterns, each drawn in code at four intensities. That is 3,872 combinations before you touch a single token.
22 colour themes
44 background patterns
Every pattern, four intensities
Exact hex values, fourteen fonts, bubble shape per side, corner radius, density, header and input. Explicit tokens always win over the preset. Two chats can share every line of markup and still wear two different brands: only the tokens differ.
A boutique hotel
A developer tool
fontFamily accepts fourteen faces
Set with one call, or one click in the designer. Changing it later takes the same one call, and every published surface updates at once.
| What changes | A static size chart | A returns dropdown form | Gnosari |
|---|---|---|---|
| Sizing before you buy | The same chart for everyone, no memory of past orders | Not applicable | Asks what they usually wear and notes the answer |
| What you get back | A shrug, or two sizes ordered to be safe | A reason code with no detail | The fit or the fault, in their own words |
| Half-told story | They just guess and hope | Submitted half empty | It asks the follow-up before it lets them go |
| Wrong item report | An email into a shared inbox | A ticket with no urgency flag | A hard stop and a person notified for anything unsafe |
| Wholesale enquiries | Buried in the same inbox as spam | Not applicable | Qualified before anyone quotes a price |
| Gets into your stack | Re-typed by hand | CSV export, if you are lucky | Signed webhook, the moment it is complete |
Everything you need to know about getting started. Can't find the answer you're looking for? Reach out to our team.
Gnosari runs the conversations an online store handles today with a size chart, a return form and a wholesale enquiry page: what a shopper usually wears before they buy, what actually went wrong when they send something back, and how big a trade order would be. It asks one question at a time on your site, then posts a typed record to your store, helpdesk or sheet, so your team acts on data instead of a free-text box.