From chaos to clarity: How I restructured a chatbot’s content for better scale
The problem: Scaling without structure
When BMO’s chatbot started growing, it grew fast. But without a clear content structure, it quickly spiraled into content chaos.
​
There were 200+ overlapping intents, making it nearly impossible for content designers, QAs, and developers to find or manage content. And when the chatbot got confused, users did too.
​
The product owner wanted to keep scaling by adding more intents and training data, believing it would reduce fallback rates. But I knew that without fixing the underlying content system, the problems would only multiply.

Challenge: Overcoming resistance
I proposed a card sorting workshop to categorize the chatbot’s intents and create a scalable structure.
The response? Pushback.
​
"We don’t have time for that. Let’s focus on training the chatbot."
​
But I didn’t back down. I knew the long-term inefficiencies were holding us back. After raising my concerns with my UX manager and director, we brought leadership into the conversation. Their support applied the pressure we needed to move forward.

Participants from the UX Retail team categorized them into 15 groups for consistency.
Solution: Bringing order to the chaos
With buy-in secured, I led the initiative to restructure the chatbot:
​
-
Facilitated a card sorting workshop → I collaborated with cross-functional teams to categorize intents, creating a clear system that reduced confusion and simplified decision-making.
-
Eliminated redundancies → By identifying duplicate intents, I streamlined content and improved tagging accuracy.
-
Established scalable guidelines → I implemented naming conventions and organized the chatbot’s content for faster, more efficient updates.
The process wasn’t just about cleaning up — it was about future-proofing the chatbot.

By the end of the project, I had laid the foundation for a more structured, scalable chatbot experience:
​
-
📈 2x increase in sprint velocity by enabling intent tagging and story breakdown by topic
-
🧠Improved NLU training accuracy due to cleaner, more consistent intent categorization
-
🧹 Redundancy removed—grouped or eliminated similar intents, reducing confusion and improving authoring
​
This wasn’t just about organizing content. It was about reframing how the team approached chatbot scale—moving from reactive content creation to a system built to grow.
The results? A chatbot that scales smarter.
2X
increase in sprint velocity (by breaking user stories into topic-based sub-tasks)
My takeaway
Scaling a chatbot isn’t just about adding more content: it’s about making sure what’s there works better, not harder.
​
By leading a collaborative solution, I turned stakeholder resistance into alignment and created a scalable system that supports both product growth and a better user experience.