Rapid-prototyping a Weather App

For this first project for General Assembly’s UX Immersive course, I created RISE/SHINE, a crowd-sourced weather app, that can to pinpoint weather on the ground.

First up, my bias was inherent before I started on the project: what else is there to talk about for weather? Locals are not likely to care about Singapore weather, as it’s humid and hot here all 365 days. I also assumed that current weather checking methods are accurate, and that users will ignore the weather in other countries.


RESEARCH

These statements were of course debunked after some user research.
The four main questions I asked were:

PAIN: What are the painful aspects of weather?
PLEASURE: What are the pleasurable aspects of weather?
CONTEXT: When and where do you deal with weather?
BEHAVIOUR: What are some words or anecdotes you will have in mind when thinking about weather?

Some other questions, like “How do you check the weather?”, were further added. My demographics were mainly 18–35 year olds, but I interviewed a 53-year-old as well, who had just gotten her first smart phone a year ago. Her answers surprisingly coincide with the rest in terms of key takeaways.

“Current apps are too general and too inaccurate.”

“The weather is unpredictable.”

“I checked the PSI hourly and daily during haze season.”

“I can plan for my usual outdoor activities if the weather’s good.”

“I check the weather to plan clothing for overseas trips.”

AFFINITY MAPPING

I was particularly interested in the group of information that told me that users liked to take control of how weather was going to do to them. SInce weather cannot be controlled, the other way of dealing with it was to change how one reacts to it.

Through some consolidating of what I’ve gathered, I was able to distill the information gathered into a problem statement:

Based on my interviews, the unpredictability of weather coupled with the inaccuracy of weather checking methods makes it difficult to plan ahead for outdoor activities.

The profile of my user, would then be who I would call, The Enthusiast.
He or she would be someone who likes to take share what is on the ground level everywhere. Whether it be outdoors, when driving during travel, my user would be excited to share infomation and hazards along the way, with other uses, building up a social network of weather enthusiast. It doesn’t hurt that this method would provide more accurate weather infomation on very specifc regions of a country.

Further reseach was done by looking at competitors’ apps, namely Accu Weather, Yahoo! and the Singapore NEW website. The app flow of Waze, a crowd-sourced transportation was also researched. Things that stood out was how the app urges users to switch on location services, for better accuracy, as well as how traffic problems pop up on the user’s current region when zooming in/out of the map.


SKETCHES

Paper prototyping commenced, and apart from a crowd-sourced weather app, the other idea I had was of a self-care app to help minimise illnessed that resulted from weather changes. At the air pollution factor of PSI300, one would be recommended home-stay and an N95 mask, and a down jacket would be recommended for an Australian winter.

USABILITY TESTING

Prototype 1 was not very well done.

The main problems were:

  • Flow is awkward from the get-
  • Location services should not be compulsory
  • (+ -) signs should be places near the map; user did not see it
  • search area not clear
  • no keyboard for typing (!)

Onward to Prototype 2, where the problems were slightly alleviated, but new problems began to surface:

  • (+ -) signs should be places near the map; user did not see it
  • Crowd-sourcing aspect of app is unclear; where to click to add
    weather and hazards?

From there on, I added in a pop-up to appear on top of the map/weather page, to urge users to contribute to the app, so as to make it a better and more accurate one. The Report button was also highlighted and allowed the user to report either the weather or a hazard. Hazard signs on the map enable users to pin-point problems enroute ahead of time.

Prototype 3 was made into an interactive one, and was tested out by a user.


NEXT STEPS

Further improvements to this app could be made by doing the following:

  • Simplify; remove unnecessary steps in the app
  • Improve login process
  • Highlight important navigation buttons
  • Test prototype on a wider range of audiences

In the three days of researching, testing and drafting prototypes, I proposed an application that would improve the unspecific forecast of other weather checking methods, and allow the user to report conditions on ground level. They can also report on hazards linked to any kind of weather conditions.

This means that via the crowd sourced method, they are able to get very accurate weather reports, that would help them in planning for any activity.

Thank you. And onward to Project 2!
One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.