Mother Hubbard

Challenge

Mother Hubbard is an AI-enabled solution that helps consumers reduce wasted food, time and money through effective grocery and meal management whilst helping businesses better understand consumer needs and drivers.

From a user perspective, the app considers household needs, dietary requirements and available ingredients to deliver relevant recipe results. So, users can discover new meals for the entire family, while reducing both food waste and trips to the supermarket.

From our client’s perspective, it collects unique insights into consumer drivers & behaviours that allows the grocery industry to understand pre-purchase aspirations with post-purchase realities – or in FMCG jargon, bridge the say-do gap. These insights help spot trends, reduce waste and optimise store layouts.

On a referral from one of our design partners, Mother Hubbard approached Sudo Roux for Mother Hubbard V2. The goal was to build on the success of its predecessor, Mother Hubbard V1, which launched as a pilot 5 years ago. With access to new technology that previously wasn’t available, now was the right time to begin a relaunch with help from Artificial Intelligence and Machine Learning.

Solution

The only way Mother Hubbard could successfully plug the gaps in both user pantries and the grocery industry as a whole was with a well-stocked database. Sowe foraged the web for the ingredients and recipes that would form the basis of our database.

AI smarts have laid the foundational cornerstone to create accurate, near-real-time sorting of data into appropriate relationships.

Utilising AI we were able to reduce the ongoing operational expenses of building a sophisticated and sizeable database, ensuring we were able to provide an enhanced user experience, catering for every need, preference and wim. More than that, the design and implementation of the AI and integration with tools such as Firebase, Webcrawler and Algolia allows for low maintenance, ongoing growth of the database that underpins the app.

Whilst the customer was from a data and analytics background, as a non-technical founder it was important that  his team were able to update data easily. To that, we created a custom dashboard for the client to easily manage and monitor recipes, the associated data and user details.

Features

The app employs a sophisticated technology stack covering web scraping, data cleaning and a custom ingredients database. A custom algorithm and automated Extract Transform Load (ETL) process from website scraping through to ingredient & recipe categorisation,  tagging, and publishing combine for a hands-off way to deliver fast, accurate and tasty results.

Mother Hubbard V2, introduces user features such as:

• Speech recognition
• Text recognition from receipts and packaging
• A user-friendly kitchen interface for managing ingredients
• Searching for a recipe by a single ingredient
• Dietary requirements
• Meal planning 
• Ratings
• Push notifications based on user behaviour & ingredient