The Weather Company
2023–24Senior Product Designer · IBM · 100M+ MAU
Overview
I was a senior product designer embedded with the IBM/Weather Company team on The Weather Channel app, a product used by 100M+ people every month across web, iOS, and Android. I led design on the Activity Hub Module and the Customer Journey experience, owned visual quality across the Andela engineering team, and worked inside the TWC Design Library to keep three platforms feeling like one app.
The challenge
At consumer scale, the cost of an inconsistent component isn't a design nitpick. It's a million people seeing the wrong thing. The Weather Channel app ran on web, iOS, and Android, with a design library most engineers had never read end-to-end. My job was the bridge: translate work from the IBM/TWC core design team into something the Andela engineering team could ship, and make sure nothing landed in production that broke the standards 100M+ users had been trained to trust.
The harder problem was the Activity Hub. Weather apps have been built around forecast tiles and radar maps for two decades. But most people don't open a weather app to read a forecast. They open it because something is happening: a thunderstorm warning, an air quality alert, a wildfire smoke advisory, a snow day. The app was good at presenting data and bad at presenting events.
The reframe
The Activity Hub wasn't a notifications screen. It was the moment the app stopped being a forecast tool and started being a situational tool.
Once that landed, I stopped thinking about the Hub as a list of alerts and started thinking about it as a triage interface: what does a person need to know right now, what's relevant to them today, and what's only worth a glance? Information density became a design problem, not a content problem.
Key decisions
Triage over chronology
Most feeds sort by time. The Activity Hub sorts by relevance: severe alerts first, then today's events, then forward-looking weather. The tradeoff: people lose “what was the latest thing?” as a mental model. We compensated with timestamps on every card and a clear visual hierarchy between active alerts and contextual updates.
One experience, three surfaces
The Customer Journey work (notifications, accounts, settings) had to feel identical across web, iOS, and Android. I worked component-by-component inside the TWC Design Library so the same setting screen on iOS felt like the same screen on Android, not a port. The tradeoff: I gave up some platform-native ergonomics (Material patterns, iOS large titles) where they would have fragmented the cross-platform mental model.
Design QA as a daily practice
Visual inconsistencies at 100M MAU compound fast. I built a habit of catching them before merge: walkthroughs of in-progress builds across all three platforms paired with a spec checklist the engineers could self-serve against. The tradeoff was time (every hour I spent on QA was an hour not spent designing) but it kept the design library honest and shrank the back-and-forth between design and engineering.
Impact
Activity Hub shipped across web, iOS, and Android as part of The Weather Channel app, used by 100M+ MAU.
Customer Journey flows (notifications, account, settings) unified across all three platforms.
Andela engineering pipeline shipped consistently against TWC design standards. Fewer rounds of design QA, fewer post-launch visual fixes.
TWC Design Library extended with new patterns from the Activity Hub work, reusable across the broader product.
Reflection
The Weather Company taught me that working at consumer scale changes what design quality even means. At a startup, quality is about taste and craft. At 100M users, it's about discipline. The willingness to hold the line on a button radius across three platforms, every week, for a year, because the alternative is millions of people getting subtly different products.
The Activity Hub work was the moment I understood that the most senior thing a designer can do is reframe a feature, not redesign it.