Design Process
For
AI FORECASTING
GET THE FULL APPROACH HERE
M&S Demand Forecasting Platform
Author: Reshma Pancholi
Project: Design Approach for Predictive Demand Forecasting Tool
Background & Context
The team are developing a data product that provides colleagues with an output showing the optimal selling price for individual clothing & home products based on demand forecasting.
Colleagues have asked that they are able to apply a number of levers to
manipulate the selling price output generated by the data product to align with
strategic goals of the business. The product will sit alongside an existing
ecosystem of enterprise systems that colleagues use to retrieve performance
information.
Task
Define an approach to designing the data product.
-
Create a plan for how you would manage and approach this project.
-
Make and state your assumptions. We haven’t given you lots of details, so use your imagination!
-
What considerations would form part of your design process and decisions?
-
Who would you work with, when, and what on?
-
Please sketch out an idea of an interaction flow and a key screen of your
choice.
Guidance: don’t spend more than a few evenings on this. We’re interested in both
your thinking process and your conclusions.
Approach
In the first instance curiosity led me to research the significance of price optimisation.
Both Deloitte and PWC confirm price matters most. In fact 60% of shoppers choose retailers based on optimal pricing. Furthermore, Analyst, Katie Paine sums up the “why” of price optimisation brilliantly, “The moment you make a mistake in pricing, you’re eating into your reputation or your profits.” And finally, Harvard studies found that a 1% improvement in your pricing can add up to 11% to your profits!
For the purposes of this assignment, these insights support and validate the case for a pricing optimisation decisioning tool but before diving into a possible design plan, I want to understand the gaps and pain points in the current eco-system of enterprise tools. What isn’t working well? And what do users really want?
Get the rest of the low down here.