Our goal is to improve the user experience for traditional menus at restaurants for customers at a restaurant. The menu has become a static item, which does not allow many clients to truly understand the dishes or variety of what's on it. We want to understand what clients who are ordering off a menu want.

Customers will use our app to explore the menu of a restaurant when they arrive at a particular restaurant. We can be presented with a different set of scenarios for each customer. Some customers may be extremely hungry and will want an efficient application, and others might be more interested in exploring the menu slowly. We want to make sure that we address both the use cases.

Here are some of the defining characteristics of the end user:

  • Any age or gender
  • Hungry
  • Can have dietary restrictions or other specific requests
  • May or may not speak English
  • May want to review food afterwards
  • Might or might not be tech-savvy

We interviewed several people ages 20-45 about their frustrations with ordering via paper menu, and what they would look for in a mobile menu.

Jacob the Runner is an athlete and programmer. He often has to deal with restricted food choices because of a string of food allergies. He is also extremely tech-savvy, and is on top of the latest trends on Gizmodo and Hacker News. He is an avid Yelp user, and often wonders why the granularity does not exist for restaurant menus. We asked him what he wanted to change on the menu. This is his direct feedback:

He is often frustrated when he doesn’t understand the terms on the menu.  He also dislikes not having control over when the waiter comes over to take his order.  He complained about the lack of an ingredients list.  He wanted to know what the food looks like ahead of time and what other users rated the food.  He would also like to be able to filter for gluten free items on the menu.

Manu the Consultant is a businessman on the go. He has often been burned by food that has tasted bad. In his life, he lacks a lot of free time, yet he has a taste for avant-garde cuisine. He would like to understand how good or bad certain items are, and how they would suit his palate. This is his feedback:

He was less concerned with how the food looks and more focused on the user reviews.  He wanted to know what are the most popular dishes and what dishes go well with each other.  He said he would write a review after he ordered his food.

Abby the Doctor is a gynecologist in a prestigious hospital. She is from Bulgaria, and often has a hard time ordering food. She is always on-call and wants her food to come fast. Also, she is a vegetarian, so she has restricted preferences. This is her feedback:

She mentioned that she has a problem with her accent when ordering food at restaurants (she said the maitré d can be condescending towards her).  She gets frustrated when her food is slow, and she would prefer to know exactly what the vegetarian fare is at a particular restaurant.  She said pictures would be very helpful, especially at an expensive restaurant where she wants to be sure she’s spending her money wisely.  Even though she acknowledged that she is 45, she said she would feel comfortable using a mobile application (she considers herself tech savvy).

Lara the Economist has been overweight for most of her life. However, she recently got into fitness, and has lost a large amount of weight. She also does not find time to cook much, so she eats out a lot. This is her feedback:

She felt that a portion size would be a huge indicator. This would help control her urges to binge. One way to do this, she suggested would be through a picture. Also, she felt ingredients would also help. A calorie indicator would perhaps be the best things on the menu she suggested, if it was not too impractical.

Chris the Mathematician is an up and coming professional. He is busy in his everyday life, and also loves to eat out. He tries various cuisine, though his favorites remain Indian or Italian. He likes to maintain control of his body and is extremely set on working when he wants to. This is his direct feedback:

He wants to know the effects of what some foods are. He also wants to know how some foods taste. Do certain foods make him feel asleep, excited, or give him a tinge? He wants to know exactly the ephemeral feeling these foods will give him.

Why is the task being done?
There is no exact way of opening up menu.io to a desired restaurant. Since the app is universal, there must be some kind of method of selecting a particular restaurant. Specifically, the client may want to search for a particular restaurant to see its menu, or to find restaurants in the area. He wants to specify.

What does the user need to know or have before doing the task?
The user can either search by name, in which case he/she should have the name of the restaurant. Otherwise, the mobile app will use location to search for restaurants nearby. The user must basically know the name of the restaurant he is at. 

Where is the task being performed?
The task can be performed anywhere, although it has special behavior if performed at the same location as the restaurant.

How often is the task performed?
This task is performed whenever the user wants to eat at a restaurant where he wants to order off of a more interactive menu.

What are its time or resource constraints?
There are no resource constraints but time might be constrained by the client’s hunger or desire to order food. Furthermore, a search is especially tilted in the favor of efficiency.

How is the task learned?
The task is meant to be intuitive and learnable, mirroring other interfaces that users will have seen if they are already mobile phone users.

What can go wrong? (Exceptions, errors, emergencies)
The GPS cannot detect the location correctly, and therefore does not display the restaurants correctly. A search might not be indexed correctly - therefore, display the wrong results. An emergency might be wanting to order food quickly, and therefore wanting to a find restaurant quickly - which is more of an efficiency issue.

Who else is involved in the task?
No one.

What are the subtasks?

Path 1:

- Check in via-location
- Presented a list of restaurants to choose from
- Select the restaurant

Path 2:

- Enter query into a search bar
- Choose the relevant text-matched name for the restaurant

Why is the task being done?
This task is the main feature of menu.io, giving the user an innovative way to explore a restaurant’s menu. The task is being done to see what items are offered at the restaurant that the client is currently eating at, and to see pictures and details of said items. The user wants to really get a good feel of what his options at a restaurant are, and the array of items available to him. He wants to explore.

What does the user need to know or have before doing the task?
The user needs to know the restaurant's menu he/she wants to view, and his/her experience might be shaped by a desire to filter the menu by criteria such as dietary constraints or food cravings.

Where is the task being performed?
The task can be performed inside of the restaurant before the client orders or while on the go as a precursor to deciding to visit the restaurant. Anywhere the user can access the internet from his/her mobile phone is a valid location for the task.

How often is the task performed?
This task is performed whenever the user wants to eat or search a menu.

What are its time or resource constraints?
There are resource constraints for data downloading/uploading and time might be constrained by the client’s hunger or desire to order food. Quick feedback to the user is essential in this case.

How is the task learned?
The task is meant to be intuitive and easily learnable. On occasion, waiters might be trained in this task in order to help the client learn the task or search the menu themselves.

What can go wrong? (Exceptions, errors, emergencies)
- The client can select the wrong restaurant.
- The client cannot browse the menu efficiently enough.

Who else is involved in the task?
A waiter can potentially aide the client in using the app, but it should mainly be used by the client.

What are the subtasks?
- Click on the relevant menu subitems
- Scroll through the relevant food items
- Click on food items for more specific information

Why is the task being done?
Certain items on a menu may be unappealing for the user, and not desirable to be seen in the menu. This task allows a user to see or show a more specific list of the items presented on a particular restaurant menu. The user wants to be able to adjust the menu according to his preferences, so that he/she can have a better viewing experience. He wants to customize.

What does the user need to know or have before doing the task?
The type of filter that needs to be applied i.e. vegetarian, gluten-free, meat-only, etc.

Where is the task being performed?
The task can be performed inside of the restaurant before the client orders or while on the go as a precursor to deciding to visit the restaurant. Anywhere the user can access the internet from his/her mobile phone is a valid location for the task.

How often is the task performed?
This task is performed whenever the user wants to get a more specific idea of the menu.

What are its time or resource constraints?
There are resource constraints for data processing and time might be constrained by the client’s hunger or desire to order food and quick feedback.

How is the task learned?
The task is meant to be intuitive and easily learnable. On occasion, waiters might be trained in this task in order to help the client learn the task or search the menu themselves.

What can go wrong? (Exceptions, errors, emergencies)
- The client can select the wrong restaurant.
- The client can select the incorrect filter, and the incorrect specifications to apply it to.

Who else is involved in the task?
A waiter can potentially aide the client in using the app, but it should mainly be used by the client.

What are the subtasks?
- Selecting the relevant filter
- Applying the filter to the relevant list of menu items

  • No labels

1 Comment

  1. Unknown User (juhokim@mit.edu)

    Problem Statement
    The problem looked a bit too generic at first, but the feedback meeting helped me pinpoint what you have in mind. The choice of a responsive mobile app is interesting. Hope you can narrow down to a more specific and novel problem in this domain.

    User Analysis
    I think the user analysis could have been better if you picked either manager or client as your main target and went deeper into that target population. Other than that, the write-up itself is good. Nice interview data, some interesting insights.

    Task Analysis
    Nice breakdown of important enough tasks. Sub-tasks are well-defined and the analysis is thorough throughout. I would like to encourage you to think more in terms of user goals that are more at the high-level than single actions. This exercise will help you weave unit actions into a meaningful process you are designing for.