Case study 06 / Commercial analytics / Synthetic-data modelling
Myntra Returns & Assortment Profitability
Independent analysis
Understanding what returns do to marketplace profitability.
An analytical model of return behaviour, reverse logistics and subsidy allocation in a synthetic fashion-commerce dataset.
Editorial illustration of the project’s focus.
01 / Context
The problem
Revenue can hide the effect of returns, repeat return behaviour, restocking and subsidy allocation on retained profit.
02 / Ownership
My contribution
Analysed return behaviour, assortment and modeled subsidy leakage across approximately 119K synthetic records.
The analytical scope
- Return-adjusted and assortment-level profitability.
- Serial-return patterns and subsidy allocation.
- Reverse-demand optimisation and restocking cycles.
How to interpret the model
The analysis explores how changing return behaviour and operating assumptions can affect profitability. Any opportunity identified belongs to the synthetic exercise, rather than a verified company result.
The business decision
Evaluate revenue alongside retained margin and the cost of serving returns. A return policy or subsidy decision needs its own behavioural assumptions and operational constraints.
Dataset & analytical context
Reading the analysis
- Synthetic dataset records
- 119,143
Data rows in the supplied synthetic workbook, excluding its header. Records are not the same measure as unique orders.
- Modelled aggregate return cost
- ≈₹35.69 lakh
The return-cost total in the supplied SQL summary. This is not an identified subsidy-leakage amount or an actual company cost.
The idea to carry forward
Revenue is not the same as retained profit. Return behaviour can fundamentally change marketplace economics.
Synthetic dataset. Not official Myntra internal data, verified company performance, a client engagement or realised company savings.