©2023 Paul Zana
Mobile App for Tracking Air Pollution Levels
(B2C, AI/ML)
(B2C, AI/ML)
Leveraging AI and machine learning, I devised a backend system that dynamically linked location air quality data, while using predictive models to fill in data gaps for locations without monitoring devices. The app provides real-time updates, empowering users to make informed decisions based on the quality of air in their current and nearest locations.
Problem statement
How might we design an experience to enable all user to access real-time air quality information for their specific location, empowering them to make informed decisions that prioritize her health and well-being.
How might we design an experience to enable all user to access real-time air quality information for their specific location, empowering them to make informed decisions that prioritize her health and well-being.
Challenge
Usability tests revealed a critical issue within the AirQo App - a misalignment between user expectations and available data. A significant 4 out of 5 users sought to find air quality information at the village or zone level, while the app could only provide parish-level data.
Usability tests revealed a critical issue within the AirQo App - a misalignment between user expectations and available data. A significant 4 out of 5 users sought to find air quality information at the village or zone level, while the app could only provide parish-level data.
Solution
Our solution involved a two-fold approach. Firstly, we proposed tagging all villages and zones within a parish to the parish-level air quality data. Allowing users to access air quality information at a granular level, aligning with their search preferences. Secondly, we leveraged AI and ML technologies to dynamically map parish location air quality data to individual villages and use predictive models to fill in data gaps for locations without monitoring devices.
Success Metrics
1. Current Location Data
2. 20% Engagement Rate
3. 20% Retention Rate
Our solution involved a two-fold approach. Firstly, we proposed tagging all villages and zones within a parish to the parish-level air quality data. Allowing users to access air quality information at a granular level, aligning with their search preferences. Secondly, we leveraged AI and ML technologies to dynamically map parish location air quality data to individual villages and use predictive models to fill in data gaps for locations without monitoring devices.
Success Metrics
1. Current Location Data
2. 20% Engagement Rate
3. 20% Retention Rate
Impact
User Growth: Rapidly growing from 567 to 5.4k users between its public beta release in the first 28 days and March 2023, it demonstrated immediate and sustained appeal.
Retention: Achieving a 50% retention rate in 2022, well beyond the initial 20% goal, signifying users' substantial reliance on the app.
Engagement Mastery: Surpassing the original 20% goal with a 322.73% engagement rate in 2022, with users engaging in 7.1k sessions, showcasing their active involvement.
Conversion: An impressive 18% overall conversion rate, with specific metrics like "Allow Location" at (0.73%), "Allow Notification" at (0.68%), "Share Air Quality" at (6.55%), and "Completing Air Quality Lessons" at (10.23%) highlighting users' active participation.
User Growth: Rapidly growing from 567 to 5.4k users between its public beta release in the first 28 days and March 2023, it demonstrated immediate and sustained appeal.
Retention: Achieving a 50% retention rate in 2022, well beyond the initial 20% goal, signifying users' substantial reliance on the app.
Engagement Mastery: Surpassing the original 20% goal with a 322.73% engagement rate in 2022, with users engaging in 7.1k sessions, showcasing their active involvement.
Conversion: An impressive 18% overall conversion rate, with specific metrics like "Allow Location" at (0.73%), "Allow Notification" at (0.68%), "Share Air Quality" at (6.55%), and "Completing Air Quality Lessons" at (10.23%) highlighting users' active participation.
Context
What's AirQo and the Air Pollution Mobile App?
AirQo is a Google-funded AI Research project — leveraging artificial intelligence to reduce air pollution throughout Africa.
Our flagship B2C mobile application seamlessly aggregates and visualizes real-time air pollution data from individual devices on the AirQo Network, providing essential environmental insights to user across Africa.
AirQo is a Google-funded AI Research project — leveraging artificial intelligence to reduce air pollution throughout Africa.
Our flagship B2C mobile application seamlessly aggregates and visualizes real-time air pollution data from individual devices on the AirQo Network, providing essential environmental insights to user across Africa.
My systematic User-Centered Design Approach
The AirQo mobile app design process began with a discovery call involving top-level business stakeholders to align with project requirements and goals. User segmentation and persona development followed, providing deeper insights into user profiles. We conducted competitive analysis and a comprehensive product audit to identify existing features and opportunities.
Utilizing the personas, we created an experience map to visualize the user journey within the AirQo Data ecosystem. Usability testing involving 5 participants yielded crucial insights into user expectations, guiding the project's direction. We refined the experience map and user personas based on these usability insights.
Utilizing the personas, we created an experience map to visualize the user journey within the AirQo Data ecosystem. Usability testing involving 5 participants yielded crucial insights into user expectations, guiding the project's direction. We refined the experience map and user personas based on these usability insights.
Test and Iterate: Post Public Beta Release
We defined problem statements and proposed solutions, advancing to wireframing and rapid prototyping. Concept testing with 5 users using Maze, an online remote testing tool, informed further refinement.
After implementation, We released a public beta — the release collected valuable usability and accessibility, achieving a 50% retention rate, 18% conversion rate, and accumulating 5000 users from Jan 2022 to Dec 2022.
Consolidating user feedback from interviews and in-app event tracking via Firebase, we refined the app while integrating the new AirQo Design System, ensuring a cohesive user experience with a focus on high-conversion features and experiences.
After implementation, We released a public beta — the release collected valuable usability and accessibility, achieving a 50% retention rate, 18% conversion rate, and accumulating 5000 users from Jan 2022 to Dec 2022.
Consolidating user feedback from interviews and in-app event tracking via Firebase, we refined the app while integrating the new AirQo Design System, ensuring a cohesive user experience with a focus on high-conversion features and experiences.
Details
Responsibility
Lead Designer
Product Strategy
Feature Scoping
Research
Interaction Design
Visual Design
Prototyping
Testing and Launch
Lead Designer
Product Strategy
Feature Scoping
Research
Interaction Design
Visual Design
Prototyping
Testing and Launch
Team
Noah N. (Lead Dev.)
Benjamin S.(Intern Dev.)
Noble M. (Intern Dev.)
Hakim Y. (Intern Designer)
Noah N. (Lead Dev.)
Benjamin S.(Intern Dev.)
Noble M. (Intern Dev.)
Hakim Y. (Intern Designer)
Date
Jun 2021 - Nov 2021
Jun 2021 - Nov 2021