The evolution of my weather app

To begin designing a weather app, it was imperative that we first interview our potential users. To do this, I started interviewing our classmates and sought their feedback. I interviewed them using the four lists method: pain points, pleasure points, context and behaviour. To follow up on this, I asked them more questions with what, where, who, when and how as well as the why why why why why method of persistent questioning. This led me to get to the very essence of their user experience and distill what was the true experience of their interaction with the weather.

Example of an interview in progress.

From this, I realised that there were a lot of notes to be transcribed. I transferred the relevant points onto post-it pads and stuck them on my room’s wall. I tried to rearrange them constantly so as to understand the essences of what the notes were telling me. We used affinity mapping to try to tease out the essences. During the course of this, I couldn’t make any sense of the numerous post-it notes at multiple points. However, I stuck to it and eventually managed to extract three salient points: 1. Users would like to have accurate weather information pertaining to the location they are interested in. 2. Users would like to be able to manage the weather information that they have received. 3. Users would like to be able to do so in a discreet and efficient manner.

Gif of my post-its

Armed with the above 3 insights, I started to think about the design direction that my app should go in. I formulated a problem statement and fleshed out my design direction based on the problem statement. I wanted my users to be an unobtrusive digital assistant that automatically incorporated weather changes into users’ schedules and informed them as needed. Thus, I set out to analyse Google Now using a flow diagram.

Flow diagram

I realised that this was far too tedious and so, I set out to design an app that would allow users to store their search history as well as retrieve their past routes quickly and efficiently. This led to the design of the first prototype on paper.

First prototype!

After consulting with Luke, I was directed to Citymappers, a navigation app that did not require users to log in. I realised that in the course of designing my app, I had failed to take into consideration the users’ privacy concerns and feelings: a user using the app for the first time would not want to give away their email address so quickly to a strange and unfamiliar app. Hence, I took this into consideration and redesigned my app’s home screen to let the user decide quickly and efficiently a route that incorporated weather changes.

Version 2

Upon further user testing, my users told me that they would appreciate the addition of current weather information on the home screen, so as to shorten the number of user interactions that they had to do. This led to the version seen below:

In summary, this has been the latest version that I have managed to come up with after incorporating users’ feedback. I realised that I need to speed up my revision cycle and improve my prototype constantly before testing it out again. I did not incorporate amendments as frequently as I would have liked and this led to the same problems being mentioned by different users repeatedly. Hence, this wasted both of our time and should have been solved by frequent revisions. I found it surprising that other apps have taken different approaches to solving the same issue: in my case, past routes. I should not have restricted myself into thinking that one always needs an account to use an app.

This app is still an interesting one to me as I have yet to find a weather app that behaves intuitively to the user. I find the question of weather to be of importance as one always interacts with the weather on a daily basis. I would like to take the idea of a digital butler further and see how far is it possible to create or design one that gives suggestions to the user and lets the user decide their schedule. This would be really helpful to the user and could really improve their time usage.

Link to the POP prototype here!

https://popapp.in/w/projects/5659f7565f14d6814ab31970/preview