Basel Alebdi & Nouf Alroqi- Data-Driven F&B Delivery: Jahez as a Leading Example| PyData Global 2023

Ai

Jahez Food Delivery: Leveraging Data-Driven Approaches to Enhance the Business, incl. Search Functions, AI-driven matches, and Personalization.

Key takeaways
  • JHAs (Jahez) is a food delivery application that uses data-driven approaches to enhance and grow the business.
  • The company has 15% of its revenue coming from search functions, and 40,000 drivers delivering orders.
  • JHAs has a data and AI department that collects search data, and uses NLP-based models to analyze queries and match them with restaurant names.
  • The company’s internal projects include data products, such as cuisine classification, and internal enablement projects, like basket analysis.
  • JHAs has a target of 44 cuisines, and tries to analyze what kind of cuisine a restaurant has by analyzing its menu.
  • The company uses multiple query checks to match search queries with restaurant names, including Levenshtein distance and spelling mistakes correction.
  • JHAs has a search missed opportunities project that aims to capture restaurants that exist in cities but are not in the JHAs database.
  • The company’s restaurant insights project provides insights about a restaurant’s orders, customers, and operations.
  • JHAs has a recommendation engine that recommends restaurants to customers based on their customer base.
  • The company uses data to personalize the application for customers, such as showcasing popular items and providing fun facts about food.
  • JHAs has a growth rate of 20% per year, and has reached more than 600,000 customers.
  • The company’s data is used to optimize the application, such as reducing the time it takes for customers to order food.
  • JHAs is exploring options to enhance its data analysis capabilities, such as using embeddings.