Too Good To Go, a marketplace designed to connect consumers with businesses to combat food waste, has identified a significant challenge: understanding why partner stores have stopped using its platform. Collaborating with Amazon Bedrock and AWS Partner Mistral AI, the company developed an analytical solution to process feedback from thousands of monthly store interactions. This initiative provided Too Good To Go with critical insights into retention issues, facilitating actionable product enhancements and a more customized approach to partner engagement.
Operating a platform that allows bakeries, grocery stores, and other food businesses to sell surplus food at discounted prices, Too Good To Go has amassed over 100 million registered users and nearly 200,000 active partner stores. The company’s efforts have successfully saved over 400 million meals from being wasted. However, the departure of some stores from the platform prompted Too Good To Go to delve deeper into understanding the root causes behind these retention issues.
The company’s existing root cause analysis processes proved to be inefficient, lacking the depth necessary to effectively tackle retention challenges. Recognizing an opportunity, Too Good To Go turned its focus to analyzing its sales team’s interactions with stores that had temporarily ceased supplying. These conversations were viewed as valuable data points that could guide product and marketing improvements.
Utilizing Mistral AI’s models deployed on Amazon Bedrock, Too Good To Go developed a solution to analyze and categorize the reasons behind stores’ departures from the platform. Amazon Bedrock, a fully managed service offering a selection of high-performing foundation models, served as the backbone for this analytical endeavor. The team extracted data from sales interactions, establishing categories for common issues and manually categorizing a subset of entries to create a training set. Through meticulous prompt engineering and iterative adjustments, Too Good To Go achieved an impressive 85–90 percent accuracy in categorization.
The choice of Mistral AI within Amazon Bedrock stemmed from its multilingual capabilities, strong performance in categorization tasks, and cost efficiency. “The main advantage of using Mistral AI models within Amazon Bedrock is that we had a lot of confidence in the system we were working with,” stated Daniel Redgate, product analytics lead at Too Good To Go. “We didn’t have to worry about data being sent to an unknown service provider to an unknown location. And we knew that if we wanted to ‘productionalize’ the model long term, then it would be much easier working within the AWS environment.”
The implementation of this analytical solution has yielded significant insights for Too Good To Go. “We’ve identified a completely new category of stores, which will require a different onboarding process and potentially different touch points with our sales team,” Redgate noted. This new category accounted for approximately 20 percent of the analyzed tickets, prompting a fundamental reassessment of the company’s onboarding strategy.
Furthermore, the analysis uncovered that many partner stores faced challenges with the platform’s scheduling feature, particularly during vacation periods, which inadvertently led to inactivity. This revelation has spurred Too Good To Go to enhance communication regarding product features and consider modifications to its user interface to improve intuitiveness. The company is also exploring new artificial intelligence applications, such as image analysis to increase transparency for consumers and optimization recommendations for partner stores, all aimed at furthering its mission to reduce food waste through technology-driven solutions.
See also
Germany”s National Team Prepares for World Cup Qualifiers with Disco Atmosphere
95% of AI Projects Fail in Companies According to MIT
AI in Food & Beverages Market to Surge from $11.08B to $263.80B by 2032
Satya Nadella Supports OpenAI’s $100B Revenue Goal, Highlights AI Funding Needs
Wall Street Recovers from Early Loss as Nvidia Surges 1.8% Amid Market Volatility
