Case Study · Fashion Ecommerce, Nigeria
Oldwyn Apparel
An AI sales assistant trained on their full product catalog — lifting the ceiling on how much they could sell by answering every product question in seconds, around the clock, in brand voice.
01 — The Situation
A growing fashion brand hitting a sales capacity ceiling.
Oldwyn Apparel sells across multiple channels — their WooCommerce store, Instagram, and WhatsApp. As the brand grew, so did inbound customer messages. Product questions, sizing questions, care instructions, availability, delivery timelines.
These were all answerable questions. The answers all existed — in the product catalog, in the sizing guide, in the FAQs. But every answer still required a human to type it.
At off-hours, questions went unanswered for hours. On busy days, they stacked up. The ceiling on sales was the ceiling on how many product questions the team could field in a day — and every question left waiting at 11pm was a sale cooling on someone else's cart.
02 — The Problem Underneath the Problem
The answers existed. They just required a human to deliver them.
The knowledge was already there — every product description, every size chart, every policy. What was missing was a system that could retrieve and communicate that knowledge accurately, in the brand's voice, at any hour.
"We had all the information. We just had no system for delivering it without someone typing it out every single time."
The business case was clear: the sales ceiling was headcount-shaped. As long as the team was the delivery mechanism, growth meant hiring — and hiring would never move at the speed of incoming demand.
03 — What Was Built
An AI agent trained on their catalog and tone.
We ingested the full WooCommerce product catalog — product names, descriptions, materials, sizing charts, prices, availability. We added the brand's FAQs, shipping and returns policy, and care instructions.
The AI agent was then trained to respond in Oldwyn's tone: warm, direct, helpful. It was configured to handle the full pre-purchase conversation — not just canned responses, but genuine answers to whatever variant of a question the customer asked.
Product recommendations
A customer describes what they're looking for — occasion, style, colour preference. The agent recommends specific products from the catalog with reasons.
Sizing guidance
The agent walks customers through the size guide, asks about their measurements, and recommends the right size — reducing returns from wrong-size purchases.
FAQ handling
Delivery timelines, returns policy, care instructions, availability — handled instantly without staff involvement.
Escalation path
For complaints or complex queries the agent can't resolve accurately, it hands off to the human team with full conversation context.
04 — The Result
The ceiling lifted. Sales stopped waiting for office hours.
The agent handles customer service around the clock. Questions at 2am get answered at 2am. Questions on Sunday mornings get answered on Sunday mornings. Response time went from hours to seconds — and the sales that used to cool overnight started closing instead.
The team moved up the stack. Complaints, unusual situations, relationship-building — the work that actually requires human judgement. The routine questions that used to set the pace disappeared from their queue entirely.
05 — The Takeaway
For any ecommerce or retail brand whose sales are capped by customer service bandwidth.
If your sales ceiling moves with how many product questions your team can answer — that is a bottleneck, not a staffing problem. The information is already in your business. The gap is the delivery mechanism, and no amount of hiring closes it at the speed demand moves.
Are sales cooling at 11pm while your team sleeps?
Book a 30-minute discovery call. We'll map where your sales ceiling is — and whether an AI agent is what lifts it.
Book a discovery call →