Top

Would you like to?

TALK TO SALES JOIN OUR TEAM

A leading UK-based retail chain wanted to assess and analyse the true return on investment from in-store promotions

Key highlights
  • Leading UK-based retail chain wanted to assess and analyse the true return on investment from in-store promotions
  • The Smart Cube was engaged for its Merchandising Analytics solution, and extensive data analytics and retail sector expertise
  • Tangible business benefits included over £10m cost savings in just one year, and £75m over four years
Download a copy

Business challenge

A leading UK-based retail chain wanted to assess the overall effectiveness of various in-store promotional initiatives and pricing changes. The business was facing information overload, with data from over a thousand stores, tens of thousands of SKUs across multiple product categories and an array of different promotional strategies. The key objectives were to:

  • Establish baseline sales to provide a holistic view of current performance
  • Estimate sales and monetary uplifts driven by various promotions/pricing changes
  • Measure true incremental gain from promotions by isolating latent factors

The Smart Cube was engaged for its extensive data analytics and retail sector expertise.

The Smart Cube solution

To generate insights that would enable more evidence-based category strategy planning, through its Merchandising Analytics solution, The Smart Cube:

  • Developed a predictive analytics tool that provides a complete and future view of expected incremental gain from in-store promotions
  • Built a customised digital platform to enable hundreds of active users (e.g. buyers, category managers, central price and promotions team, space planners), to assess and analyse promotion performance at all levels in the product hierarchy
  • Created automated data pipelines for weekly refresh and updated predictions to ensure near real-time insights for end users
  • Provided recommendations and insights to improve future strategies around instore promotions and pricing initiatives

The core engine for the solution is driven by advanced statistical and machine learning models, deployed across all SKUs. A wide range of potential attributes were considered to build a comprehensive set of models, including store features, space allocation, price, promotion attributes, events, and seasonality. Further modules were added to give statistical estimates on pull-forward and cannibalisation, thereby providing a true and more complete view of promotions effectiveness.

Results

The client has derived tangible business benefits including:

  • Over £10m cost savings in just one year and £75m over four years
  • More accurate pricing for multiple SKUs
  • Elimination of all ineffective promotions

Since deploying the analytics tool, the client has been able to:

  • Effectively and correctly assess true performance of in-store promotions
  • Evaluate data at multiple levels, from granular SKU level, to category and store levels
  • Identify which promotions work for which SKU/category (e.g. ‘half the price’, ‘save x%’ or ‘save £x’)
  • Create evidence-based promotional strategies based on these findings
  • Arrive at the optimal discount percentage for each product

Value delivered

  • Unique approach to assessing accurate ROI: The analytics tool is one-of-a-kind in the market, using a unique holistic approach to isolate latent factors which generates a true measure of promotions ROI
  • Contextual intelligence for multiple users: Through an end-to-end customised solution – leveraging best-in-class technologies and advanced analytics – users could access timely, relevant information to address their challenges
  • Actionable insights for strategic decision making: The analytics tool has the capability to answer the most complex questions pertaining to promotions and provides insights to improve forward planning
  • Strategic impact: The business has been able to identify the right promotions to implement at the right time and devise promotional strategies with positive ROI, increasing the top line