Can technology innovate the age-old restaurant menu?

‍

Machine learning, or ML, has the ability to create unique, personalized menus for each patron. Machine learning, or ML, has the ability to create unique, personalized menus for each patron.

‍

By stating what your allergic to or/and what your dietary preference is, an ML algorithm can eliminate meal options, substitute ingredients, take ingredients out, or create a new meal option.

‍

You can expand the scope to include all menu ingredients and allow patrons to deselect options based on their taste preferences too.

‍

The Benefits

An ML menu will make the dining experience more efficient and unique for the patron and restaurant.

‍

It could take less time for patrons to order, attract new patrons, ensure health and diet preferences are upheld, and eliminate the back and forth between the patron and restaurant (servers and kitchen).

‍

How It Works

A web based application using ML is trained on what every diet consists of and what ingredients are in each meal at the restaurant.

‍

For example, vegans can’t eat cheese. Anything with cheese is eliminated from the menu.

‍

This might eliminate a whole menu in some instances. So, the restaurant can tell the algorithm which meals or ingredients they are willing to substitute or eliminate.

‍

For example, a restaurant could provide vegan cheese as a substitute or eliminate eggs from pad thai.

‍

Another option for the algorithm is labels. If a restaurant makes vegan or gluten free options, it could label it for the ML algorithm to quickly identify.

‍

‍

3 Use Cases

When you’re in the reservation flow, a new page appears. You enter your dietary restrictions and allergies to get an email sent to you with a personalized menu.

‍

If you’re vegan, you will see the vegan meals, and any non-vegan options that can be customized to be vegan.

‍

Not everyone makes reservations and not every restaurant accepts reservations. While browsing the menu online, you can filter based on diet, allergies, and if applicable, ingredients.

‍

We can’t assume everyone is going to make a reservation or look at the menu beforehand. A final point of implementation is at the table.

‍

A hostess could deliver tablets to the patrons with the menu application loaded. The patron uses the same filtering process to find their meal.

‍

A unique twist could be allowing the patron to order at the restaurant by connecting the application to the restaurant's POS system. This would eliminate potential manual errors from the waitstaff and free up their time to do other things.

‍

Conclusion

An ML menu takes the best of high-end dining experiences and mixes it with the seemingly limitless options of Chipotle or Sweetgreen.

‍

In a world full of options, ML menu’s limit options in a positive way, while providing a unique restaurant experience.

‍