Makani Tech: Build. Automate. Evolve.

Case studies · D2C skincare

D2C skincare brand: COD orders confirmed before they ship

About 207 hours a month of calling and "where is my order" chats automated

Mumbai, Maharashtra, India · Tier 1 city15 people, 400 orders a day
saved a month
207 h
saved a month
team time a month
₹93,000
team time a month
fewer returns if RTO drops 7 points
17 a day
fewer returns if RTO drops 7 points
to build
3 weeks
to build

About this case study: it is based on a business typical of the ones we work with, not a named client, and every number comes from the assumptions at the end of the page. Swap in your own and the sums still work.

The problem

Of 400 daily orders, about 240 are COD. Some buyers cannot be reached, some ordered on impulse, some gave a wrong address. Each return costs shipping both ways.

The support team calls a share of COD buyers to confirm, and answers around 60 "where is my order" chats a day by checking the courier site.

What we build

  1. 1

    Confirm on WhatsApp

    Every new COD order gets a WhatsApp message with Confirm and Cancel buttons, plus an optional small discount for paying online.

  2. 2

    Ship only what is confirmed

    Confirmed orders go to the shipping queue. Orders with no reply in 12 hours become a single call task.

  3. 3

    Tracking without asking

    Courier updates are sent at dispatch, out for delivery and delivered.

  4. 4

    Order status answered

    An AI assistant answers order status questions from the Shopify order and the tracking data, and hands everything else to the team.

Before and after

StepBeforeAfter
Confirmation callsAbout 96 a dayOnly unanswered orders
"Where is my order" chatsAbout 3 minutes eachMost answered automatically
Address problemsFound at deliveryCaught at confirmation

Where a person stays in control

Refunds, cancellations after dispatch and complaints always go to a person. The assistant only reads order and tracking data.

Running cost

Roughly ₹15,000–25,000 a month for messages, the support tool and AI usage, mostly driven by order volume.

Tools

  • Shopify
  • WhatsApp Business API
  • Shiprocket or courier tracking
  • AI assistant
  • n8n

How the numbers work

  • 240 COD orders a day; 40% were called at 3 minutes each ≈ 4.8 hours a day × 30 days = 144 hours.
  • 60 status chats a day at 3 minutes = 90 hours a month; 70% answered automatically ≈ 63 hours.
  • Team time costed at ₹450 an hour: 207 × ₹450 ≈ ₹93,000.
  • If confirmation lowers COD returns from 25% to 18%, that is about 17 fewer returns a day on 240 COD orders. Measure your own rate before and after.

Estimates, not promises. Use the savings calculator with your own hours and costs.

Want this for your business?

A free 30-minute call is enough to tell whether it fits, what it would take and what it would cost.

More case studies

Shopify apparel brandLos Angeles, California, USA

Apparel brand: an AI assistant for order, return and sizing questions

About 52 support hours a month back, with refunds still handled by people

Most support tickets ask the same three things: where is my order, how do I return this, which size should I get. The team answers them one by one.

saved a month
52 h
saved a month
support time a month
$1,300
support time a month
of tickets resolved without a person (assumed)
~50%
of tickets resolved without a person (assumed)
to build
2–3 weeks
to build
Read the case study
Textile wholesaleSurat, Gujarat, India · Tier 2 city

Textile wholesaler: WhatsApp orders that type themselves

About 178 hours a month of order typing and dispatch calls taken off the desk

Dealers send orders as WhatsApp messages, photos of handwritten lists and voice notes. Three people retype them into a sheet and then into Tally, and spend the afternoon answering "has my parcel left?"

saved a month
178 h
saved a month
staff time a month
₹62,300
staff time a month
to check an order, down from 12
2 min
to check an order, down from 12
to build
3–4 weeks
to build
Read the case study
All case studies