MS E Dynamic Markdown

Introduction:

This module recommends, on the basis of each style's/ SKU’s ongoing performance and stock status if we should continue the current discount on the style/ SKU is working for it or if we should increase or decrease the discount on it to improve its sales or margins respectively.

Business use-cases:

It is capable of handling business cases like -

In-Season discounting (Business as usual)
Event discounting like - End of season sale, Platform events, etc.
Old merchandise liquidation
Identifying styles breaching guardrails, and many more...

Impact delivered:

Up to 6% margin improvement on fast movers for online point of sales
71% improvement in ROS and a 10% discount increment for low-performing styles in offline stores
2x increase in the frequency of decision making
Successfully handled liquidation scenarios for in-season and old-season scenarios.

Parameters:

The algorithm takes into consideration all the important parameters to make decisions -

The True Rate of Sale ^TM - i.e. ROS of a style when it was present in a healthy size set to compare true performance amongst styles
The Ageing of style - for how many days has the style been trading for / how old is it
the stock cover of the style - how many days of stock do we have for that style and then also
the Health of the style - if the style is currently available for sale in a healthy size set or with only a few sizes options left.
Sellthrough has been added as another option for offline brands that prefer this over inventory cover.

Algorithm Flow Chart:

Features:

The module course corrects the recommended discounts to always meet the company-level discount targets by analyzing the ROS and discount relationship for every category.

The algorithm also takes into consideration the price elasticity factor which ensures that no further increase in discount is recommended after the increase in discount has stopped having any positive impact to increase the rate of sale (the module learns from the previous discounting cycles, and improves the quality of decision based on it).

Custom grouping of styles for analysis allows multiple groups within the same category (e.g. collared t-shirts and round-neck t-shirts can be analyzed separately).

Flexibility to run at different levels and groups of stores or channels (e.g. separate online and offline demand).

The module has been built with a lot of flexibilities that can accommodate any kind of fashion brand as it takes brand-category (and more AG) level guardrails to ensure it acts within the brand's allowance/expectations on discounts.

The module can work at any desired frequency.

Style-level discount overrides can be given.

MS Enterprise Public