A Premium Shoe Manufacturer Leveraged A Contextual AI Asset To Run Real-Time Marketing Experiments And Drastically Reduce Time To Insights






Impact Measurement Tool

Developed impact measurement framework to measure real-time impact of 100+ experiments run across marketing, pricing, and CRM functions

Cost of measuring the impact was reduced by more than 80%

Experiments were recommended based on historical results and this enabled decision makers to perform smarter trials

Assisted the revenue management team to plan sales and in turn, production by developing a tool that could churn out accurate forecasts across multiple levels

Built market mix solution to measure the impact of various marketing activities and optimize spends across channels which resulted in incremental marketing impact of 6 percentage points                                 

An upcoming premium shoe manufacturer wanted to increase the brand’s effectiveness and outreach. While they were the fastest growing brand in the niche given their unique, eco-friendly products, they were relatively new to the market and wanted to further improve the brand presence. The existing decision-making process was decentralized, and it was challenging to identify activities generating high ROI. The time taken to arrive at decisions is key, and the client wanted to speed up the decision-making process to achieve high-impact results at pace.

TheMathCompany developed an impact measurement tool that could measure impact of experiments/ changes implemented in near real time and suggest changes to the experiment, if necessary, thereby, speeding up the decision-making process as well.

The optimum level of data granularity had to be identified for the tool. Once functional, the tool would identify scenarios where lift measurement was essential, provide insights into real time impact and recommendations on experiments that would help in improving operational efficiency.

To identify activities that are generating favourable results:

  • The granularity of tests was cross-referenced with data and key performance indicators were identified.
  • Appropriate algorithms were utilized for forecasting and developing accurate metrics.
  • User flow and wireframe for the product were defined.
  • After extensive functional testing and user acceptance testing, the product was deployed on a cloud platform.
  • Upon assessing the features of the cloud platform and taking the real-time usage into consideration, data pipelines and applications were developed.

The resulting tool comprised a multi-layer architecture:

  • A presentation layer that comprised user access, input and output visualization
  • A model runner where statistical models running to measure the impact could be altered as and when the need arose
  • The application layer, i.e., an interface connecting the user interface and backend applications
  • A data layer made up of input data with a provision to store results
  • When using the tool, first, measurement KPIs and granularity of measurement had to be identified. Then, parameters had to be defined for events whose impact had to be measured. A control strategy determining factors like geography, had to be chosen for the event. Upon setting up these filters, impact could be assessed, and the results could be visualized.

Related resources

Subscribe to Authentic Intelligence – our monthly newsletter

Stay up to date with the latest marketing, sales, and service tips and news.