THE CHALLENGE

Getting kids to commit to nutritious choices can lead to a healthier lifestyle. Schools have the opportunity to foster good habits early on. But accommodating picky eaters and dealing with outside influences, such as the media, family, and peers, can be daunting.

Design an experience where schools can influence students’ food choices when they’re on campus. 

A Bit of Context

I don’t have children, and the last time I was one was around the time Kurt Cobain was topping the charts. The most tech we had in the 90’s was when the teacher rolled the tv into the room to watch Bill Nye, so I was stoked to dig into this whole new world of science fiction today’s kiddos call a school day.

Outside of this challenge sounding super fun, I ultimately chose this prompt because I am a total and complete outsider without bias.

My goal was to dig into actual problems from real people and design an elegant solution they would find valuable.

UNDERSTAND

Understanding the problem space is *the* most important step in designing anything. I go by the double diamond model, where I first diverge (learn everything) then converge (focus on specific problems) before digging into solutions. A breakdown of how I caught myself up is as follows.

1. DISCOVER (diverge)

1. Talk to people (experts and end users)

2. Immerse myself in material (articles, science journals, podcasts)

3. Look at current solutions to some of the issues uncovered, form opinions about said solutions (what’s working, what’s not?)

2. DEFINE (converge)

1. Build themes and clusters

2. Define the problems

3. Create “how might we’s”

4. Product brief

Talk to People

I kicked off my user recruiting with a very sophisticated tool I like to call “Facebook”. I shot it out to the universe, hoping after a few conversations I would find a thread or two to follow as I aimed to define my problem focus.

Turns out I know a pediatrician, a few teachers, and a handful of moms who are struggling with picky eating for a variety of reasons. I called them up and started with a pretty baseline battery of questions. You can see a couple early sessions here.

The conversations were pretty open ended, which allowed for me to sit back and allow curiosity to knock when they touched on issues I could dig into a bit deeper later.

Below are some examples of how my brain started to sort some of their comments into nuggets of opportunity and constraint.

DESK RESEARCH

I was absolutely floored by the conversation I had with Alana; I had no idea parents with autistic children experienced such a struggle around mealtime. The numbers become interesting when I learned 1 in 68 children are diagnosed with ASD, with 80% of those children having problems with selective eating. I was very curious to see if there were any school programs that helped address this issue, and keen to see if I could build a solution that could potentially make nutrition education more inclusive.

I was also very interested in this “eat a rainbow” idea I learned from my pediatrician friend because it touched on making nutrition education easier for parents and caregivers by chunking information into manageable (and fun) concepts. These became my two main points for further reading.

MOST VALUABLE INSIGHTS

• There are several Rainbow Programs that currently exist, but they are not easily accessible to teachers. However, these programs have a wealth of resources (videos, worksheets, lesson plans) and follow school standards, which could serve as a great source of content, reducing the need for me to design an entire program for a school to accept and adopt.

• Increased exposure to whole foods early on in a child’s development improves likelihood of choosing those foods later in life. Often the source of picky eating stems from not being exposed early enough, often enough (at least 8-13 times), or coupled with enough positive reinforcement.

• There is a lot of overlap in therapy techniques for picky eating and sensory food aversion. Both neurotypical and ASD children could stand to benefit from sensory food aversion treatment and new food exposure in a classroom setting around the age of 4-6 (pre-K and kinder levels).

• Sitting in on a workshop around treating sensory food aversion, I discovered all variables of food preference and exposure are gradable (e.g., texture, flavor, degree of exposure). Anything gradable can be connected to a feedback model and recommendation algorithms, which makes this area attractive for designing a solution around behavior modification.

• Children are more likely to try and accept new foods when they see their peers doing it. This makes the classroom the ideal place for food exposure programs.

Competitive Analysis

I looked at a wide variety of nutrition education apps/sites, and in this case, making a table of features was really just a waste of time. However, there are two types of products on the market I would like to call out to illustrate my choice in direction.

Ultimately what we are trying to achieve in this challenge is to inspire children to make healthy eating choices, which at its core is behavior modification.

Each solution below demonstrates a unique style of habit development.

Solution 1: Kurbo

Kurbo is catching some heat at the moment for being what some are calling “a weight watchers app for kids”.

Initially, I was drawn to their crafty use of stoplight colors meant to teach kids “no” foods, “slow” foods, and “go” foods. It felt like an elegant solution for food tracking (those colored dots make it really easy to monitor and categorize!), but the rightful controversy lies in the foundation of its style of behavior modification.

It is based in shame.

This isn’t a new concept. In fact, I learned my own sister was put on a similar program when she was a kid. Did it help her? No. But what it did do was make all of our family dinners extremely tense, and it resulted in a lot of food secrecy and embarrassment.

In addition, Kurbo’s stoplight quick view model does not take color blindness into account. It is very difficult to track the most important cues from small dots when you can’t see the difference very easily.

Side note - who is the user? I’m not entirely sure this solution is kid friendly, but I digress.

Solution 2: Smoking Cessation Apps

Below is an app called Smoke Free, but the market is flush with a number of apps designed to help people stop smoking with a similar program style. The system they are employing is, at its core, Cognitive Behavioral Therapy (CBT). It focuses on changing people's reactions to their urges to smoke by changing thoughts and behaviors in the moment.

By establishing a reward model around good behaviors instead of punishing bad ones, we encourage a higher likelihood of program adherence, which naturally results in greater habit forming/breaking potential.

CBT is the most effective form of behavior modification for avoidant and fearful behaviors (in our case, picky eating due to sensory aversion) because it establishes positive emotions around good food experiences rather than promotes shame around “bad” foods.

I put all of this info in my back pocket and shifted gears into phase 2 of the understand phase - defining the key problems and developing the “how might we’s” to solve them.

Defining the Problem

After all of that reading, I came to some conclusions as to where a lot of the issues exist that could potentially benefit from a tailor made, school supported food education experience.

1. Kids develop food aversions at an early age because we don’t expose properly or frequently enough, and are inconstant between school and home.

2. Schools and parents stick to “safe” foods they know children will like/won’t be allergic to because of ease and lack of communication (between parents + schools and caregivers + children).

3. Caregivers often reward the wrong things (using food as reward) and use negative reinforcement to deter “bad” food choices, which promotes shame and negative associations around food we want them to eat.

4. Caregivers incorrectly assume that failure to eat food to completion (clear is he plate!) is an entirely failed attempt, and are quick to label food to avoid.

5. Many parents have no idea what their children are eating at school because schools rely on students to report their day to them. It is difficult for teachers to keep ALL parents abreast of their student’s food exposure progress.

How might we…?

• How might we … expose kids of varying sensory aversion levels properly to food in a way they understand, feel safe doing, and feel good about after?

  
  

• How might we … make sure we’re exposing children often enough during these impressionable times (whiles reducing feelings of defeat in parents and teachers along the way)?

  
  

• How might we … make it easy for parents, students and teachers to grasp the Rainbow and “exposure eating” concepts?
  
  

• How might we … make it easy for caregivers to track student progress and make future exposure decisions?

PRODUCT BRIEF

After I fully define the problems I want to focus on, I find it helpful to write up a list of design principles in the form of a product brief. It keeps me on track and allows me to make decisions more easily when I’m in the “diverge” phase of solutioning, which is coming up next.

1. The product will be usable for teachers, parents, and students, with a focus on pre-k and kinder levels.

2. The product will focus on rewarding exposure, not punishment for choosing the wrong foods, NOR punishment for opting out of eating.

3. The product will be inclusive for both neurotypical children, as well as children with Autism and severe sensory food aversions.

4. The product will establish consistency between school and home food exposure practice.

5. The product will provide an easy, lightweight way for teachers to communicate a student’s progress to their parents.

6. The product will be intuitive to use and follow for children aged 4-6, and will be accessible to children who are non-verbal.

7. The product will enable teachers to first activate the student’s schema before launching into eating the food by providing additional age-appropriate teaching material.

8. The product will be compatible with mobile devices, tablets, and smart boards.

9. The product will integrate with Google Classroom, leveraging GC features as much as possible.

EXPLORING SOLUTIONS

Next, it was time to take all of these great questions and constraints and begin ideating an ideal experience. This stage is imperative for helping shape abstract ideas into more concrete tasks.

I like to start big and whittle down, making sure all of the ideas are valuable, executable, and ultimately solve the identified problems in a simple, elegant way.

1. Develop (diverge)

1. Determine vision and strategy

2. Establish user flows and jobs to be done

3. Create an IA that makes the most sense to the users

4. Sketch UI to review and iterate

2. Execute (converge)

1. Settle on hi-fi prototype to test

2. Iterate

3. Build (not shown here…)

4. Repeat (not shown here…)

The Vision

Rainbow is an app-based program that will enable teachers and parents to establish and track positive food exposures, allowing them to communicate student progress in an easy, lightweight way so they are better positioned for success.

The Strategy

Working closely with teachers and established Rainbow-based programs, we can ensure school adoption and inclusion of age-appropriate content enrichment.

IMPORTANT FLOWS

First and foremost, it is important to design the high level foundation of a behavior modification experience before anything else. I decided to employ CBT techniques, focusing on reducing food anxiety by multiple exposures, coupled with positive reinforcement (reducing pressure, reward for trying). As “it takes a villiage” for these things, I chose to actively involve the parent in this system. It looks a bit like this. (Pinch to zoom!)

Taking into account that the product will be used in the classroom, and must, above anything else, get full teacher support, I sat with teachers Katherine and Amber to design a theoretical positive food exposure program they would be interested in implementing in their lesson planning.

TASK CLUSTERING + JOURNEY MAP

I started grouping some of the tasks we envisioned teachers, students and parents would need to complete in order for this tool to operate successfully (see graphic below), and put them into an experience map. In early sketches, I included student tasks, but for simplicity am noting the only direct interaction children will have with the tool will be for the 1:1 feedback sessions.

Visualizing this flow and distilling into a task list helped me take my very abstract ideas and begin to develop more concrete interactions that need to be considered.

SIMPLIFY THE ASKS

After reviewing each of these tasks, I began to cut for the following reasons:

1. Some can be done by the system itself. The less cognitive load you can place on a user, the better the experience.

2. Some features were pretty neat ideas, but ultimately didn’t add to the core value of the tool itself. These features could be explored in the future as value adds.

3. A few of these asks would need a some edge case consideration. For example, some parents might not take the time to fill out the evaluation form. Making it easy for a school nurse to submit allergies to ensure food safety might help in those situations.

TRY DAY: GRADED FOOD EXPOSURE

Borrowing from the workshop I took with Marsha Dunn Klein, I designed the “Try Day” food exposure challenge along a graded continuum, from least intrusive up until eating the food in its entirety.

I applied a color of the rainbow to each of the steps, so by the time the student gets to the end, they earn a full rainbow.

The colors would be reflected as progress states for the children to follow along as they collect all of their colors. I’ll dig into other ways I integrated the rainbow into my designs in later sections, but imagine each of the challenges depicted below in the “Try Day” step to follow this line of thinking.

SKETCHING + IA

This is the messiest, most iterative part of my process. Ideally, I don’t touch a pixel until I have a pretty fair idea of how things are generally laid out and what the flows look like, working very closely with users to make sure it makes sense in the moment.

How I’ve broken down some of the most important aspects to noodle on:

1. Onboarding (teacher vs parent)

2. Lesson Planning

3. Scheduling

4. Full Class Try Day Challenges

5. 1:1 Feedback sessions

6. Tracking and Future Recipe Recommendations

1. ONBOARDING

• What kind of information do we need collect at the onset to power the magic in the app’s backend? Who can provide it?

  • Allergies, sensitivities, food likes/avoids + food qualities likes/avoids

  • Some of these things can be taken from school records

• How are users onboarded? Is it a link sent from the teacher given to parents? Is it emailed? Can they use Google Classroom to bulk invite?

  • Ideally Google Classroom can manage as much as possible

  • When teacher emails bulk from GC, code is generated for easy class assignment

• What information can we provide the users about Rainbow early on so they know what to expect and how to best use it?

  • This is where we can really teach parents the proper methods for food exposure and how to easily see food types via color profile

  • Goal setting is key to frame expectations

2. LESSON PLANNING

• How do teachers plan which foods to feature on which week for each month?

  • All foods are filtered for allergies/sensitivities and give a good mix of likes and dislikes for all children.

• What content types are the most valuable to them and how do we show a vast content library that will have multiple image thumbs, titles, etc.?

  • Video, digital books, printables, activity ideas

  • Everything can be fed in from other Rainbow programs, perhaps connected via a Google Classroom app for teacher social share

3. SCHEDULING

• How can we make it intuitive for parents to sign up to volunteer food for Try Day at the beginning of every month?

  • The app will automatically send a reminder notification to parents (email, Google Classroom, mobile app push)

  • Parents will select recipes from a curated list for easy decision making. All recipes will detail exactly how to prepare and for how many children.

• How can we make sure teachers are able to see who is designated to bring food on specific days, and send reminders when not all of the slots are filled?

  • In-app and Google Classroom Sign Ups overview, push notifications if its getting close.

4. FULL CLASS TRY DAY CHALLENGES

• What exact levels should we take the students through that would reflect best practices for stepped food sensory exposure?

  • See above rainbow levels

• How can we set up the interactions for a teacher to be able to cast the flow on a smart board and still be able to make selections or see private notifications on the tablet or smartphones?

  • Quick little nudges to make sure ALL students are taken care of, if some of them are falling back, the system will be able to tell (whereas a teacher might not be able to easily)

• How can we make sure the UI is physically accessible on the smart board, as teachers are interacting directly with it?

  • Try to keep all of the most repeatable actions low (how big/high is the average smart board?)

  • Perhaps give the option to switch location of frequent actions in settings

• How can we make the UI understandable and exciting at a pre-k level so students can follow along and WANT to be included in getting rewarded?

  • Make sure imagery can support teacher explaining things like “fuzzy like a sweater” or “hard like a rock”.

  • Put eyeballs on some things. It makes it fun and friendly.

  • Always give motion and audio feedback when a positive behavior has been completed. Pavlovs dog can attest.

5. 1:1 FEEDBACK SESSIONS

• How can we make sure students are not distracted by UI elements we don’t want them to interact with and entice them to the ones we do?

  • Keep the elements we don’t want them to touch FLAT and unexciting. Make the elements we want them to touch skeuomorphic.

• Make sure the feedback sessions mirror the full class experience so students know exactly what is expected of them and how to interact with the elements.

• What kind of words make the most sense for describing the characteristics of food that can be used as a shared language between student and caregiver?

  • There is a lot of testing and research that can inform this.

  • Make sure there is a limited number of options on the screen (6-8) so as not to overwhelm the students.

6. TRACKING & RECOMMENDATIONS

• What tracking information would parents be most interested in seeing and in what format? (real time diary vs. stat overview)

  • How many times a student has been exposed to a food

  • Evolving taste profile (crunchy, soft, sweet, etc)

  • Daily achievements (so they can commend them on it when they get home)

  • Overall compliance - are they doing well? Improving?

• How can the exposure feedback sessions influence a recipe recommendation algorithm?

  • Every recipe is rated on the following scales, which can hide or show recipes and group them into categories

    • Texture

    • Flavor

    • Color

    • Temperature

    • Smell

    • Previously liked or disliked

    • Allergies/sensitivities

7. REWARD SYSTEM

• One star is given for each step taken during the Try Day challenge. One rainbow is given for each fully completed challenge.

• Class Dojo is the go-to model for teachers who like to use real time reward behavior. Rather than reinvent the wheel, for this challenge, I’m just going to call it out as the best in class scenario for Rainbow’s reward element (sans the demerit feature).

HIGH FIDELITY PROTOTYPE

I wanted to make sure I was able to fit in at least one round of iteration from feedback before calling it a day, so I picked the most important aspect of this experience to focus on needing to be visually correct: the 1:1 feedback screens.

As I don’t have super easy access to pre-k aged children, I relied, again, on my teacher users to help steer me in the right direction.

I showed them the screens below, explaining the scenario and asked if they thought it would be easy for children (both verbal and non-verbal) to understand and interact with.

For the most part, they loved it. They did have a few points of feedback, so I went in and made some tweaks.

1. Show visuals of each sense (e.g., hand by touch, eye by see), and make it more central to the options

2. Group the text within the button so it looks like one object

Below is what I call “a very casual remote user review” and “working with what you got in the time allotted”. But hey, it helped!

Final Designs

After shopping this concept around to more teachers and a few moms, describing all of the bells and whistles I was thinking the experience could include, I asked them very bluntly…

“What value stands out the most to you?”

Hands down, they all agreed on two points.

1. “It makes it SO easy for the teacher to bring this program into their classroom and give feedback to parents about their kids’ progress without having to do a bunch of extra grading and reporting. They can focus on actually teaching and helping the students, not all the busy work.”

2. “Parents can take all of the feedback from the classroom and actually bring it into their kitchen without having to hunt for recipes they think their child might like. The fact that the recommendation engine gets smarter is such a great motivator for parents to keep using it after the program has ended.”

FINAL THOUGHTS

There is so much to explore with this concept, and I kind of wish this wasn’t just a fun little design challenge. I might share it with some of these experts I’ve been stalking, though. I’m so curious as to what they might think!

If I were to take it further, however, after hearing what excites people about it the most, I would focus on those two big value points as my north star. Slim it down to what it needs to do and be and get rid of the rest.

And of course, test it with children and teachers in an actual class setting. That would be so fun.

Thanks for reading <3

xoxo Lauren