Designing for clarity
Redesigning the information architecture of an app as complex as Intercom is no easy feat – here's how we did this to allow for future growth.
Every product eventually faces a moment when its foundation begins to strain under the weight of growth. For Intercom, this moment came when our information architecture (IA) – the framework organizing features, settings, and workflows – could no longer keep pace with product development.
The warning signs were clear: users struggled to find basic workflows, settings were scattered everywhere, and navigation labels created confusion. When one user described our product as “wandering through a maze of features,” we knew we had to act.
In a new essay, our Senior Product Designer Pranava Tandra dives into the highly detailed process of reimagining the Intercom information architecture. Check out the full article here, or read on to discover some of the key lessons we learned.
Rethinking our approach to clarity
Rebuilding your IA is like renovating a busy train station without stopping the trains. To tackle this complex challenge, we anchored ourselves to a single principle: clarity. But what does clarity actually mean in practice?
For some users, clarity meant fewer paths to the same destination. For others, it meant dedicated spaces for new tools. Through user interviews and testing, we discovered that true clarity isn't about what makes sense to us as builders – it's about what makes sense to our users in their day-to-day use of the product.
This revelation transformed our approach. We stopped thinking about IA as a design exercise and started seeing it as a fundamental principle: removing friction so users can focus on their goals, creating frameworks teams can build on, and laying the groundwork for graceful scaling.
Here are three crucial lessons we learned about achieving clarity in product design.
Design for real people, not edge cases
While we thought grouping AI tools into one section made perfect sense, early tests showed users weren't clear what “AI” meant in practice. The lesson? Real people don't use products in perfectly predictable ways. Instead, they have diverse needs, assumptions, and goals that we need to account for. Our job is to design for their reality, not our idealized scenarios.
Simplicity takes effort
If simplicity were easy, building every product would feel effortless. Instead, it's messy and iterative, full of trade-offs and dead ends. We debated labels, reworked structures, and tested countless prototypes. Each iteration brought us closer to clarity, but also taught us that simplicity isn't a starting point – it's the hard-won result of relentless focus.
Clarity is a moving target
Redesigning something as big as your IA isn't something you “finish.” It's a living system that evolves with your product and users. What works today might need adjusting tomorrow. This redesign was a crucial step forward for us, but we’re fully aware that it won't be our last. We'll keep listening, learning, and iterating as our users continue to use our product.
Check out Pranava’s full post to learn more about how her team approached this challenge and the specific improvements they made to our information architecture.
What’s coming up
We have loads of exciting startup events happening over the next few months. Here’s what’s coming up:
🇺🇸 San Francisco
Y Combinator meetup with Solidroad (YC W25)
Thursday 20th February at 6 pm
Hackathon: More details coming soon
Saturday 22nd February
ODF meetup
Wednesday 5th March
Startup Grind speakeasy: More details coming soon
Tuesday 29th April
🇬🇧 London
Lovable meetup: Go from Idea to App to Customers in One Hour
Tuesday 11th March at 6 pm
If you’re interested in attending any of them, check out this post from our Startups Senior Manager Alan Mc Glinchey and reach out to him directly on LinkedIn.
What we’ve been up to
Making the best of a bad situation: Lessons from an Intercom outage
Imagine starting your day with a page about elevated exceptions, diving into Datadog, and uncovering a lurking 32-bit integer limit in one of the most critical parts of your app’s data model. 😱
Well, that’s what happened during a particularly chaotic, and unusual, outage for Intercom. What followed was a five-hour marathon incident response involving monkey patches, migrations, and feature flag gymnastics – but despite the stress, it was certainly educational.
Speaking at Rails World, Staff Product Engineer at Intercom Miles McGuire shares more about what went wrong, how the team fixed it, and the lessons they learned to prevent it from happening again.
He also dives into some of the technical details the team implemented after the fact to make Intercom’s Rails app more resilient.
If you’re curious about how to avoid similar pitfalls – or just enjoy tales of debugging under enormous pressure – check out Miles’ presentation.
Calling all founders 📣
Join our Early Stage program to receive a 90% discount on Intercom's AI-powered support platform – which includes Fin, our game-changing AI bot capable of resolving half of your support volume, automatically. You’re in good company.
From the archive
Intercom product principles: How we focus on delivering outcomes
At Intercom, success isn't measured by what we ship – it's measured by its impact. As our Co-founder Des Traynor explains: “It's what you ship versus what happens because of the thing you ship.”
This piece from our VP of Research and Data Science Karen Church digs into how we think about delivering outcomes as well as outputs to drive real results for our customers.
We’re driven by these three core principles:
Define outcomes first: Before building, our team clearly outlines desired customer and business impacts, thinking about everything from how much time we want to save to what our revenue growth goals are.
Measure everything: Every feature we ship has comprehensive analytics tracking, allowing us to measure the impact the feature has on user behavior.
Iteration is essential: Shipping is just the start. We actively drive adoption and conduct regular outcome reviews, iterating until we achieve our defined goals.
Ultimately, we believe that great features alone don't build successful companies. What really counts is delivering measurable value for both customers and your business.