The client is one of the largest Fast-Moving Consumer Goods Company with a heritage of over 80 years. Its products include food and beverages, cleaning agents, beauty and personal care products.
In order to evaluate the effectiveness of sales promotions, the client makes use of Nielsen marketing data and compares it to their internal sales promotion master sheet on a monthly basis. Since the data alignment is different in these datasets, it becomes a tedious and time-consuming process to extract the required information. As part of this process they,
- Manually cross-reference multiple spreadsheets, search for all misspellings and abbreviations, search for other languages and branding localization, filter the best matches for actual products desired, manually extract the relevant data.
- Waste hours in simply extracting the information to do the job of mapping Nielsen marketing information to internal sales promotion master.
Therefore, manual activities tend to be error-prone ending up at an accuracy of 35 – 40%.
In the current increasingly challenging and competitive environment, businesses rely heavily on promotions and offer to generate higher returns. Though, promotions are powerful tools for enhancing sales for many businesses, evaluating the effectiveness of Sales Promotions has been a challenge. It consumes a significant amount of man-hours to analyze and process the enormous unstructured information, which resulted in a delay in the process of benchmarking the sales performance.
The objective is to provide a robust solution coupled with cognitive technologies for their Sales Promotion Analytics.
In order to overcome the above challenges and accelerate the business processes, JK Tech Smart Analytics developed a Sales Recommendation Engine that is built using a Natural Language Processing (NLP) Platform. The solution uses 2 AI neural networks:
- The Language Model.
- The Abbreviation Model.
These models collectively compared the Nielsen marketing data and internal sales promotion master data to automate the matching exercise with increased accuracy and reduced Total Cost of Ownership (TCO).
In addition, the recommendation engine enabled them to understand their competition in the market and take the necessary actions to improve their sales and customer retention.
Automating the Business Process:
- Increase data accuracy to 90%.
- Reduced the TCO (Total cost-of-ownership) by 70%.
- The solution saved time and money by automating the complex sales performance benchmarking business processes and tasks.
- Enabled faster business decision-making by leveraging cognitive technologies like Artificial Intelligence & Machine Learning.
- Increased productivity and operational efficiency by optimizing the workforce.
- Increased value by identifying and maximizing sales opportunities.