Cover Image.jpg
 

Initial Concept

The core idea of this project was an app that would play music or audiobooks that would fade out gradually to help the user to fall asleep, and then slowly fade in in the morning to gradually wake them up. The user would catch Zs (points) when they got a good night’s sleep and earn badges for hitting new milestones and building healthy habits.

 
Storyboard 1.jpg
 
 

SCOPE & STRUCTURE

To figure out which features to include, I used the MoSCoW method to prioritize them and figure out which were essential to create the minimum viable product. This involved listing out every feature I was considering including in the app and then placing them into columns marked Must, Should, Could and Won’t , to prioritize the features and determine which were needed to create a minimum viable product.

Once I had narrowed down the features, I organized them into a flow to start figuring out the structure of the app.

MSCW V3.jpg
Flow V3.jpg
 

WIREFRAMES

Once I had the flow, I started sketching out wireframes to look at the screens I would need to create and to figure out how they should be connected to each other.

Wireframe V2.jpg
 

MOOD BOARDS

I created several moodboards and ultimately narrowed it down to these two to determine the styling of the app.

Griffin1 - Serene.jpg
Griffin5 - Soft.jpg
 

STYLE GUIDE

Style+Guide.jpg
 

VALIDATING RISKY ASSUMPTIONS

Although the timeline for this project was pretty short, I wanted to make sure my design was still as effective as I could make it. For that reason I decided to conduct a pretotype experiment to validate my riskiest assumption. My main concern with the design so far was that it might not actually improve the user’s sleep quality. For my experiment I decided that if at least 25% of my testers showed some sort of improvement I would count it as a success.

To conduct the pretotype experiment I gave each of my testers a sleep quality scale to fill out which gave each of them a numerical score. Over the course of the next week I gave them audio clips that would fade out after about half an hour to play as they were falling asleep. At the end of the week I had them fill out another sleep quality scale so I could compare the scores to see if there was a change.

Risky Assumptions 1.jpg
 

testing core functionality

One Night Stand Pretotype

60% of my testers showed an improvement in their scores which was a lot more than I had hoped for when I started the experiment.

If I were to perform this test again I would probably adjust a couple things. None of my testers were able to test for the full week so ideally I would have liked to have a longer time period to conduct the experiment to make sure the data was not a fluke. Along the same vein, a couple of the testers mentioned that they had a hard time remembering to play the audio before going to bed so having reminders in place would improve the quality of the data. Overall though, the experiment results were enough to alleviate my worries about the basic function of the app.

Pretotype 1 Results.jpg
 

CREATING A NON-GUI PRODUCT

For the next stage of the project I began designing a non-GUI product that could work with the app. This would be a smart speaker that would sit on the user’s bedside and remind them to go to sleep on time and read them stories to help them fall asleep while also serving as a phone dock.

 
Storyboard 2.jpg
 
 

VALIDATING RISKY ASSUMPTIONS

My biggest concern with this part of the project was that people may not like the styling of the smart speaker. To test this, I created a model and then posted an ad along with a landing page for it to gauge interest.

Risky Assumptions 2.jpg

CREATING THE MODEL

PINOCCHIO PRETOTYPE

process.jpg
Rip holding cord no text.jpg
 

GAUGING INTEREST

FAKE FRONT DOOR PRETOTYPE

Ad & Landing Page.jpg

Pretotype results

After advertising for five days to a broad audience, the link on the ad had been clicked a total of 27 times. The website only listed 25 visitors coming from Facebook but both of those numbers surpassed my original goal of 15 clicks. Although both the ad and website could be improved, it worked as a proof of concept to show that there was interest.

Pretotype 2 Results.jpg
 

THE FINAL DESIGN

The final design consisted of a wireframe done in Adobe XD as well as a physical model of the smart speaker.

The app includes a leaderboard so you can compete for Zs and badges with your friends as well as a settings section where you can set your bedtime preferences.