While at Vyana, we were approached to create an evidence-based baby development tracking app. My challenge: make milestones interesting for a user, and gather enough data that an AI could be trained to make predictions about a child's subsequent development.

RESEARCH

We knew we were targeting new parents, but who are they? I spoke with a number of new parents, and conducted an online survey to find out more.

We needed to find out what they are looking for from a "baby" app?

MARBLE

​Team:

Lead UX Designer,  Illustrator

Michael Mabee

UI Designer
Rozy Penaloza

Animators

Joong Jin Park, Gav Patel

Developers

Tintu Jacob & Will Morgan

Business Analyst

Hema Jeyaraj

Scrum Master

Shobana Chandrasekharan

Content Creation & Marketing

Stuti Dutt, Emily Kennaley & Dalal Kheder

Medical Science Liason

Michael Papaioannou

Product Owners

Phedias Diamandis & Nardin Samuel

Tools Used:

Figma (Low-fi's)

Adobe XD

Adobe Illustrator

Available:

mymarble.ai
Apple App Store

Google Play Store

A couple of interesting findings: 

  1. quite a number of my respondents either lived with or were regularly supported by extended family

  2. there was a certain amount of fear related to doctors visits with baby

  3. for most respondents online sources of baby health information ("Dr. Google") were their go-to, but were not wholly trusted

  4. social media was not generally viewed in a positive light

I looked into the wide range of other baby apps that are already on the market. After downloading and exploring their functionalities, I found that while they are generally successful at what they do, none of them tackle more than one facet of of what a parent might be looking for from a baby development app: tracking, milestones, photo sharing, development info, shopping or community.

I found a number of interface structures (like hamburger menus and top-of-screen navigation) that stemmed from being web-first products, that would hinder a one-handed, baby-carrying user on mobile.

In looking at the UI, I found that the apps created by institutions tended toward a trustworthy but institutional look, while those that focused on shopping and community had a brighter, more exciting look to them. In creating a app that was intended to be informative as well as visually engaging, we were going to have to find a comfortable middle ground.

PLANNING

In building the app, I would have to figure out how to string together a product around a skeleton of milestones, if we were going to be able to feed the AI enough data. It simply wasn't enough to keep users coming back, though.  We needed to give them a great deal in return.

 

We would provide users with solid, evidence-based information. We'd give them a number of "hooks", like access-limited photo sharing, that would keep parents coming back. We'd find ways to accommodate all kinds of families, including core and extended family members, and multiple babies. We needed to find a way to present a doctor with data in the form of a comprehensive report, to back up the things parents were seeing happening with their baby.

DESIGN & TESTING

From my own experience having children, I knew we would have to create an experience that would allow users to dip in and out, giving them small bites of information tied to their child's age so it's always relevant. That became the Home screen.

 

We would need a place to allow users to do their own exploring, when they had the time and presence of mind to do so. That became the Library.

 

They would need a place to get feedback about their baby's growth as related to the milestones we were creating. That became Milestones.

Throughout my time with children in daycare, there was always a certain amount of data gathered about feedings and diapering, used as a measure of how the child was feeling that day. That became the Tracking tab.

In testing the low-fi prototype, users found it easy enough to discover the functions of the app, and to navigate around. More detail was needed to show them the full scope of interactions, and the depth of information at their disposal, ie. articles with source links, graphs that would show their baby's progress, etc. It was time to move to high fidelity.

The high-fidelity prototype helped us define Marble's personality, and add complexity to the ways the user would interact with it. With input from the team's BA and developers, we filled in a good number of the gaps in the initial design, and began development. With the devs getting to work, I built out nearly 500 illustrations for the the first two years' milestones.

In testing early APKs of the app, it became clear that users needed some hand-holding on account setup, so we added notifications related to completed tasks. While adding those, we extended them, adding timed reminders, new age group, and weekly progress report notifications so they would have external triggers to keep coming back.

 

We also decided to surface all of the interactions from other people in the parent's circle: showing photos added by other users in the home screen, and completed milestones with the user-added photo in place, so they have both an additional status indicator about that milestone and a cute permanent record of the accomplishment.

OUTCOMES

We launched Marble into the market on November 23rd, with happy product owners. We're actively promoting it, and waiting to see how the market responds. I already have a number of possible extensions plotted out for MVP2, and am eager to continue.

Visit mymarble.ai for more information, or the Apple App Store or Google Play to download it now.

© 2020 Michael Mabee