Vrinda Store: why revenue fell 20%

Why I picked this question

"Sales are down" was a sentence I heard often in sales, usually with no detail on who stopped buying or where. A total alone doesn't tell you what to fix. I wanted to break a revenue drop down until it pointed to a specific group of customers the business could actually act on.

How I approached it

1. Loaded the 31,047 order rows into SQL and checked the data before analysing it.
2. Used SQL to break monthly revenue down by sales channel, product category, gender and age group.
3. Compared March (the peak) with November (the low) for each group to see which ones fell the most.
4. Built a one-page Power BI dashboard with filters so a manager could explore it themselves.

What I found

1. Revenue peaked in March and slid for the rest of the year

Monthly revenue went from Rs.1.83M in March to Rs.1.46M in November, a fall of about 20% (around Rs.363K a month). It levelled off in December.

2. The drop was concentrated, not spread across the business

• Channels: most of the decline came from Amazon and Myntra, two of the three biggest channels.
• Products: Sets and Kurtas, the two best-selling categories (Rs.10.5M and Rs.5.0M), fell the most.
• Customers: spending by women dropped 22 to 30% across every age group, while spending by men dropped only 5 to 8%.

3. Women are the core of the business

Women brought in about 64% of all revenue. So when their spending fell, the whole store felt it. The rest of the business was fairly stable, with a 93% delivery rate and an average order value of Rs.744.

What I would recommend

1. Run a win-back campaign for female customers on Amazon and Myntra, focused on ethnic wear (Sets and Kurtas), where the drop was largest.
2. Test it before rolling it out. Send it to one group first and compare against customers who didn't get it.
3. Track revenue by gender and channel every month so a drop like this is spotted in weeks, not at the end of the year.

I estimated that bringing back part of the lost spending from this group could recover around Rs.2M a year, about 10% of total revenue. This is an estimate based on the 2022 numbers, so the test in step 2 would show the real effect.

What I'd look at next

The data doesn't show prices, promotions or competitor activity, so I can't say why women stopped buying. Next I would look at whether prices or discounts changed on Amazon and Myntra during those months.