Outsourced support teams
Answering services
Unified CCaaS platform Voice, chat, QA, and workforce management — unified. fashion & apparel
AI agents guide shoppers to the right size before they buy, handle styling questions in their language, and process returns and exchanges without friction.
The short version: put AI on restock dates, order status and first-pass size questions; put a person on “it shrank”, “it looks nothing like the photo”, anything tied to an event date, and anyone with an audience. AI voice is US$0.15 a minute, chat US$0.15 a session, a shared human agent US$300 a month for 500 minutes, and we take accounts from half a seat to 200+. We normally have an account staffed within five days. And the frame we work from: almost every fashion complaint is an argument between three documents — your size chart, your listing and your care label.
how it plays out
Customer comparing sizing across brands
Customer wants outfit inspiration before committing
Customer questioning material quality after delivery
Sold-out item from latest collection
the team behind it
real members of the global team Fashion-Trained Specialists
65+ Languages
Follow-the-Sun Coverage
the part nobody prices in
A customer who writes in about a garment is rarely describing the garment. They are reporting that something they read did not match what arrived. The size chart said medium. The listing said soft and relaxed. The label said machine wash cold. One of those three was wrong, or the customer read it differently from how you meant it.
That matters because the three documents have different owners. The size chart belongs to your product team and the listing to marketing. The care label, under the FTC Care Labeling Rule, is attached by the manufacturer or importer, who must state on it the regular care the garment needs. A support team that files all three as “return — did not suit” throws away the only place in the business where the three get compared.
So what we sell here is narrower than a returns desk. We answer size and fabric questions from your documents, say which document the answer came from, and tag each complaint to the document it contradicts, so the pattern reaches someone who can fix it. We have done this kind of work for an American fashion brand selling into China, running its local numbers, its service-level reporting and its exchanges. What we cannot do is make a size chart accurate. What we can do is make sure you hear about it before the season is over.
who takes which call
Most vendors sell you either an AI receptionist or a live answering service and leave the allocation to you. Since we run both, here is the split we would actually recommend — including the calls we think should never touch automation.
| The call | Route to | Cost of getting it wrong |
|---|---|---|
| "What size should I get? I am a medium in most brands." | AI | A parcel shipped twice and a sale refunded once |
| "When is this coming back in stock?" | AI | A customer who buys the lookalike somewhere else |
| Order status, return label resend, exchange tracking | AI | A second message, then a chargeback |
| "It shrank. I washed it exactly like the label said." | Human | A label problem filed as a customer mistake |
| "The color is nothing like the photos." | Human | A listing that keeps producing the same return |
| "I need this for a wedding on Saturday and it is the wrong size." | Human | An event that cannot move and a customer who will not forget |
| A creator with a large following complaining in your DMs | Human | A reply written in private and screenshotted in public |
what it costs
Most vendors in this category quote by form. These are our published packages — not a quote, since language mix, hours and volume all move the number. The per-minute figures are simply the monthly price divided by the included minutes.
| Option | Published price | Per minute | What it's for |
|---|---|---|---|
| Live shared team | from US$650 / month | includes 1,500 min → ≈ US$0.43/min | Several storefronts and marketplaces sharing one team through a season change |
| AI voice | US$0.15 | per minute | "Is this back in stock?" and "when does the sale end?" asked by phone |
| Web chat | US$0.15 | per session | Size questions on live chat, Instagram and WhatsApp while someone is still deciding |
| Live shared team | from US$300 / month | includes 500 min → ≈ US$0.60/min | A new-collection week or a sale weekend, when messages arrive faster than your team can read them |
| Dedicated agent | from US$5.80 | per hour, per agent | A named person for VIP clients, press and creators, and anything that may be posted publicly |
Anything involving a person is a starting price. Two things move it: which languages you need, and how senior an agent the work actually requires — a bilingual agent handling refund authority is not the same hire as one taking bookings. AI voice and web chat are flat, because there is no hiring in them. And every model above is open to every business on this page: which one fits is a question about the shape of your volume, not about your industry.
Minimum engagement is half a seat — not a five-seat block. We also bill per resolved issue, from US$1, where you pay nothing if it is not resolved; that model fits businesses where a customer problem actually gets solved on the call, not businesses where the call is intake.
do the arithmetic yourself
Pull a month of return and complaint reasons and put each into one of four piles: the size chart was wrong, the listing was wrong, the label was wrong, or nothing was wrong and the customer changed their mind. Only the last pile is a cost of doing business. The other three are defects, and they repeat at the rate you sell the style.
For each style in the first three piles, multiply its returns by what one return costs you: shipping both ways, the item you cannot resell at full price, the time to process it. Set that against the cost of fixing the document — a fit note, a new photo, a corrected label from your supplier. We are not quoting an industry return rate here; your own return reasons, sorted this way, are the only number that decides it.
The reason this belongs on a support page is that the sorting happens on the support line or not at all. A return form with a dropdown collects “did not fit”. A conversation collects “I am a medium everywhere else and this was tight across the shoulders”. The first is a statistic. The second is a size-chart fix.
setup
Five days is our normal window to staff a live account — that is five days to find and place an agent who fits your business, not five days of software setup. AI voice can go live sooner because there is no hiring step.
your existing software
Amazon through the SP-API, two-way with Shopify, plus eBay — so an agent handling a case can see the order, the shipment and the return without asking the customer to repeat the number. WhatsApp, LINE and TikTok come into the same queue as email and chat. If a platform is missing from that list, ask before you assume: we will tell you straight whether we connect today or whether it needs building.
your existing software
Amazon 
Shopify 
eBay 
WhatsApp 
LINE 
TikTok Running something else? Tell us what it is and we will say straight whether we connect to it today or whether it needs building.
what comes off the call
Most of this already lives in your product and merchandising files. Collecting it is what lets an agent answer a fit question from your data instead of from general fashion advice.
The fit-notes line decides how useful we can be. An agent reading a generic size chart gives generic advice; one reading your fit notes can tell a customer that this style runs small.
where AI should not be
"It shrank, and I followed the label."
This is either a faulty garment or a wrong label, and neither is the customer’s fault. A bot that sends a return code closes the ticket and throws away the only evidence you had that a label needs fixing.
An order tied to a date that cannot move.
A wedding, a graduation, a trip. The customer does not want a policy; they want to know what can still arrive in time. That needs a person who can check the real options and be honest about the ones that will not work.
Size conversations that turn personal.
Pregnancy, a change in weight, a body the chart does not describe. These are conversations where tone matters as much as the answer, and a scripted reply reads as not having been listened to.
A complaint from someone with an audience.
A creator or a journalist writing in private may be drafting in public. The reply needs to be one you would be comfortable seeing screenshotted, written by a person who knows who is on the other end.
a model we're not selling you
Time to process a return is the number most returns desks lead with, and a fast, polite return is genuinely good service. As a service level for a fashion brand, though, it measures the handling and never the cause.
A return on a style whose size chart runs small can be processed in minutes, and so can the next one on the same style. The number improves while the same document keeps producing the same return, and a team measured only on speed has no reason to ask which document the customer read.
So we tag each return with the document it contradicts, next to how fast it was handled, and put a person on complaints that point to a chart, a listing or a label. Any model on our price list is still yours to pick, including per resolution for order status and label resends, where the answer in the conversation is the whole job. What we will not do is call a quick return a solved problem.
when we're the wrong choice
questions
AI voice is US$0.15 per minute and web chat is US$0.15 per session. Live agents start at US$300 per month including 500 minutes — about US$0.60 per minute — or from US$5.80 per hour for a dedicated agent working only your account. The minimum engagement is half a seat, not a five-seat block.
Five days is our normal window to staff a live account — that is five days to find and place an agent who fits your business, not five days of software setup. AI voice can go live sooner because there is no hiring step.
Yes, on the same number, with agents in our Bolivia and Nicaragua operation centres — native Spanish, not machine translation. English and Spanish share one queue, and we cover 60+ more languages if you need them.
That is the normal setup. You decide which call types go to AI and which reach a person, and you can change the split as your volume changes. Nothing about that decision is locked in at signup.
The label is not a suggestion. The FTC Care Labeling Rule, 16 CFR 423.6, requires manufacturers and importers to attach care labels to textile wearing apparel, and says those labels must state what regular care is needed for the ordinary use of the product. So when someone followed the label and the garment still failed, the useful question is not what the customer did wrong but whether the label was right. We log it against the style and its label, not just as a return. Ten of those on one style is a label problem, not ten careless customers.
They advise from your size chart, your fit notes and your past return reasons, and tell the customer which of those the advice rests on. If a style runs small and the listing does not say so, we will tell the customer and flag it to you the same day. What we will not do is invent a fit claim your product team has not made.
Yes, with one limit that decides how useful we are: each channel has its own return rules, and some of them are not yours to change. We keep each channel’s rules in the intake and tell a customer plainly which process their order follows, rather than promise an exception that neither of us can grant. A marketplace return follows the marketplace, even when the customer wrote to you.
start
We will show you AI-guided sizing and styling in action with your catalog.