Mayank Enterprises
Staff spent half of every day answering the same questions about stock and delivery, and customers still waited up to 90 minutes for a reply.
support workload cut
- WhatsApp Business API
- Trigger.dev
- n8n
- OpenAI
Where they were
A consumer electronics retailer selling across a storefront and WhatsApp, where most customer contact arrives as a message rather than a call. Every message was answered by a member of the shop floor team, between serving people in front of them.
What was broken
The overwhelming majority of those messages were three questions: is it in stock, where is my order, and can I return it. Answering them by hand consumed about half of every working day, replies took 45 to 90 minutes at busy times, and outside the 10 hours somebody was on shift nobody replied at all.
What we built
A WhatsApp agent trained on the live product catalogue, so stock answers come from inventory rather than from memory.
Order-status lookups answered directly in the thread from the order system, with no human in the loop.
Returns initiated conversationally: eligibility checked, the request logged, and the customer told what happens next.
Anything the agent cannot resolve escalates to a person with the full conversation attached, so the customer never repeats themselves.
Round-the-clock cover, which is where a meaningful share of the volume turned out to be — messages sent after closing no longer wait until morning.
What it did
Measured after go-live, over 2 months.
- Total support workload removed
- 70%
- Queries resolved with no staff involvement
- 65%
- Time to first reply, from 45–90 minutes
- <30 sec
- Cover, from 10 hours a day
- 24/7
"Our team used to spend half the day answering repetitive stock and delivery queries. The WhatsApp engine now handles nearly 70% of customer chats automatically."
This build is our e-commerce & retail proof point. See what we build for e-commerce & retail.
Book a Free Strategy Call