The Problem
AT Kearney is a global management consultancy with a team of forward-thinking and collaborative experts who deliver immediate, meaningful results and a long-term transformational advantage to clients. AT Kearney Procurement & Analytic Solutions has grown into a service practice delivering large strategic engagements leveraging advanced analytical techniques for better results. The company came to S2DS to develop a proof-of-concept application for an innovative approach to pricing: an automated, autonomous and statistically rigorous solution used in a live, online sales environment for optimal price setting.
The Data Science Project
The team consisted of four S2DS participants with analytical PhDs in collaboration with a mentor from AT Kearney. The team developed a flexible simulation environment in which to test, assess and refine the pricing solution against a limited test-bed of real-world pricing scenarios. Applying concepts from experimental design, statistical testing and advanced regression techniques the team was able to create a pricing tool that successfully discovered optimal prices. Furthermore, through rigorous design testing and improvement, the final pricing tool was able to significantly outperform benchmark approaches for optimal price discovery – both in terms of transaction volumes required and financial benefits realised during price discovery.
We were impressed both by the calibre of S2DS participants and the results achieved in such a short span of time. (Alex Romanenko, Director, Head of Analytics at AT Kearney UK)
The Solution
The proof-of-concept work conducted by the S2DS data scientists successfully showcased the potential in this approach to deliver tangible, competitive pricing advantages to AT Kearney’s clients. Building on this strong foundation, AT Kearney is now developing this innovative pricing approach further, to increase its applicability and robustness to a wider range of real-world scenarios with a view towards testing the solution itself in a live client environment.