top of page
DishLogo.png

Dish

iOS APP - BRANDING, UI/UX DESIGN

Dish is a clinically approved diet self-management app for people with Irritable Bowel Syndrome.

HOVER TO REVEAL

INTRODUCTION

In late 2018, we built an iOS app prototype to validate the hypothesis that IBS patients want affordable end-to-end dietary support. This involved digitizing the low-FODMAP medical diet in a way that minimizes the risk of user error, in order to qualify for clinical approval and alleviate the dietary anxiety that often haunts IBS patients.

MY ROLE

User research, conceptualization, branding, product design, user testing

THE TEAM

1 front end developer, 1 GP, 1 clinical dietitian

TIMELINE

Jun 2018 - Dec 2018

Discover
Group 1.png

To create fertile ground for problem discovery I kicked off with desk research and user interviews. The problem statement fell out of that, and was followed by a systematic solution brainstorm with the wider team.

USER INTERVIEWS

A painful problem

The most common IBS pain points were physical pain and discomfort (100%), frustration around the inability to socialize (89%), and anxiety around ‘accidents’ (72%). Motivation to alleviate painpoints was extremely high, based on an average annual IBS spend of GBP 1,120 and an average of 9 solutions tried.

“I don’t care if I have to eat cornflakes for the rest of my life, I’ll do anything to get symptom free.”

Inadequate low-FODMAP dietary support

Only 12% of interviewees had sought clinical dietary support, citing GPs' unwillingness to refer them, long waiting lists, and/or not being able to afford private care. 83% Frequented online support communities, notably Facebook groups. While these did a good job of meeting IBSers’ emotional support needs, only 8% felt they gave them adequate support in self-administering the low-FODMAP diet.

"Doctors don't take us seriously. At least [on Facebook Groups], we can connect with each other."

The low-FODMAP diet: a blessing and a curse

Considering the low-FODMAP diet’s broad adoption, sentiments were surprisingly negative. Key pain points were its complexity (96%) and the vast amount of time spent on meal preparation (92%).

“The [low-FODMAP] diet makes my symptoms worse: the anxiety about getting it wrong gives me flare-ups.”

“All these blogs and brochures talk about what I can't have. Why don’t they tell me what I can have?”

PROBLEM BREAKDOWN

🧐 The NHS only provides antidepressants

Responsible for research, conceptualisation, design, user testing and evaluation of clickable iOS app prototype. Responsible for research, conceptualisation, design, user.

⚠️ Self-administration constitutes a health risk

💸 Many patients can't affod the 1,200 for private dietary care

Costs GBP 1200. Responsible for research, conceptualisation, design, user testing and evaluation of clickable iOS app prototype. Responsible for research, conceptualisation, design, user testing.

🎉 Placeholder

Responsible for research, conceptualisation, design, user testing and evaluation of clickable iOS app prototype. Responsible for research, conceptualisation, design.

PROBLEM STATEMENT

While IBS patients want to do the low-FODMAP diet to alleviate their symptoms, the diet's complexity can worsen symptoms and professional support has limited access. How might we make low-FODMAP guidance simple and affordable?

SOLUTION IDEATION

We used a cluster chart to systematically think through the ways in which we might make low-FODMAP guidance simple and affordable.

We decided in favor of a dietary guidance app, because the standardized nature of the diet lends itself to digitization while its potential for scale could make it a big market opportunity. Additionally, digitizing a medical protocol allows us to explore infrastructure for cost-effective medical care which could be an interesting strategic direction for future growth.

Cluster chart example - Color.png

Solution ideation cluster chart made in Lucidchart

FEASIBILITY CHECK

To find out if our 'digital dietitian' was likely to pass clinical approval checks, we flushed out the logic model and jumped on a call with a real dietitian.

We built the logic model based on two key features:

  • The anxiety around 'getting it wrong' could be addressed by sending notifications every step of the way

  • Grocery home delivery would be the most impactful way to reduce the vast amount of time spent on meal prep

 

The clinical dietitian confirmed that our model matches the accuracy of the brochures she currently gives patients, given one technical adjustment ("before starting reintroduction, confirm that the patient hasn't had any symptoms in the past 5 days").

Flowcharts - Algorithm flowchart example.png

Low-FODMAP diet flowchart made in Lucidchart

She also said that a low-fodmap guidance app makes a lot of sense to her (whoop!), but suggested it may be a dietitian support rather than replacement tool. "We're complementary: while your app can’t assess a patient’s broader health picture, I don't have time to send my patients daily dietary instructions or place low-FODMAP grocery orders for them."

Selling into dietitians was outside of our mission of providing affordable guidance for those who can't afford a dietitian. However, we decided to include dietitians in our traction tests as this may teach us something about future growth opportunities.

Define
Group 2.png

With tentative clinical support in our pocket we felt confident to start defining the MVP. We flushed out which assumptions the MVP should test, what the MVP should look like, and how we define success for each assumption.

ASSUMPTIONS

To test if IBS patients are willing to pay for a digital dietitian, we listed our assumptions across the 4 types of product risk and sorted them from most to least risky.

We chose to focus our first traction test on our 3 riskiest assumptions, covering core value, accessibility and viability risks.

Screenshot 2023-04-23 at 3.51.43 PM.png

Excerpt of product risk assumptions table

THE MVPs

To test these assumptions we decided to launch a two-fold MVP: a landing page with waitlist signups to check if there was a commitment to a digital dietitian, and a clickable prototype with a survey to see if our product delivers on key needs.

Landing page with waitlist signups

We chose a Dropbox-style waitlist MVP because email addresses are a small but real form of commitment. This MVP would speak to the high-level validity of our value proposition and give us a rough idea of how much demand we can capture at a low cost (assumptions 1, 3).

Prototype survey with selected follow-up calls

To validate the usability and value of the digitized guidance itself, we will develop a clickable prototype paired with a survey (for quantitative insights), paired with 1:1 observational calls with our early adopter group (to understand the underlying 'why'). 

We will also put the prototype in front of a few dietitians to ensure medical compliance and gauge their sentiment (assumption 2).

PERFORMANCE INDICATORS

Since I had no baseline data to inform what good would look like, I decided to use general statistics to loosely define worst, base, and best case outcomes.

The base case KPI for the landing page's unique visits was established as (total social media audience) * (0.3 for active users) * (0.5 for impressions reach) * (0.06 -> for Reddit's average impression click-through) = .. . This calculation may be erroneous in many ways, so we'll take results with a grain of salt.

The prototype survey KPI's were based on Sean Ellis' product-market fit survey, which suggests that a product has reached market fit if at least 40% of survey respondents say they feel "very disappointed" when asked how disappointed they would feel if this product no longer existed.

Screenshot 2023-04-23 at 5.30.34 PM.png

Scenario-based KPIs for our two MVPs

Develop
Group 3.png

Confident we're building the right next thing, I kicked off MVP design. I mocked up the wireframes and a minimum viable brand, and turned them around into a clickable prototype and a landing page. To avoid scope creep I time-boxed the work to one week.

LOW-FIDELITY WIREFRAMES - 1 DAY

Based on commercial requirements (customer pain points), clinical requirements, and ethical requirements (minimize risk of user error).

Lo-fi wireframes.png

Sample of lo-fi wireframes made in Figma

BRANDING - 1 DAY

To avoid a sluggish design process, I decided to put together a high-level brand identity upfront. Since  branding wasn't key to test our initial assumptions, I time-boxed the work to 1 day.

Deliverables: we need a rough brand identity for product and visual design, covering:

  • Name and logomark

  • Key colors (brand and greyscale)

  • Typefont (system font suffices)


Requirements: based on user sentiments the brand should communicate reliability and simplicity


By clearly defining minimum deliverables, using hard-cap timeboxes and leaning on references from existing brands, I was able to create a high-level brand identity in a day.
 
Hi, Dish! ❤

Group 573.png

Minimum viable brand made in Figma

HI-FI WIREFRAMES - 1.5 DAYS

I rolled the brand out over the low-fi wireframes and referenced market-leading apps (e.g. Deliveroo for food delivery) to save time while making sure we followed best practice UX. Stand on the shoulders of giants, and fine-tune from there kind of thing.

Screenshot 2023-05-07 at 5.56.45 PM.png

Sample of hi-fi wireframes made in Figma

LANDING PAGE MVP - 0.5 DAY

CLICKABLE PROTOTYPE - 1 DAY

Deliver
Group 4.png

DISTRIBUTION

Landing page distribution:

  • Facebook groups
    Reddit forums
    Selected academia
    GP surgeries

✈️ Distribution

We decided to test a broad variety of free distribution channels, and adjust our focus as we learned what was effective. Initial distribution channels included:

​

  • Direct to patient

    • Facebook groups (70 groups with +- xk followers)

    • Reddit channels (10 threads with +- 20k followers)

    • Printed leaflets at GP practices

  • To dietitian

    • Cold outreach (+- 100 email addresses)

    • PR in specialty publications (+- 2-5 mags)

RESULTS

Landing page distribution:

  • Facebook groups
    Reddit forums
    Selected academia
    GP surgeries

bottom of page