PLATFORM
TIMELINE
TEAM
After launch, Booth Selection saved organizers significant time managing exhibitor booth registration. Setup ran across three steps: upload floormap, customize booth tiers, and place pins on the map.
The first two steps worked well. But for events with hundreds of booths, placing 50–200+ pins by hand became the step organizers were most likely to abandon.
For organizers
Existing users spent significant time on tedious pin placement; others gave up before ever seeing what Booth Selection could do.
For business
Exhibitor Registration is one of the platform's major revenue streams; slow setup capped growth from existing users, and drop-off held back new adoption.
Yes, but what could the detection realistically handle, and where would it break?
Incorporating AI was a decision made from day one. So my job at this stage was to figure out where detection would break, what edge cases would come up with real event maps, and how the design should handle them.
TWO INSIGHTS SHAPED EVERYTHING I DESIGNED AFTER
01
Maps have different level of information AI could work with
Maps with color- or size-coded booths can suggest groupings. Maps with only booth outlines gave it nothing to work with.
02
Detection accuracy varied based on the map input
Number detection was ONLY reliable at high resolution; Size-based grouping only worked when the booth sizes were clearly different.
03
Maps often contain items AI can detect but can't judge
Some maps included booths that weren't for sale; Non-booth items like stages and info tables, also have number.
With the principles set, the next question was how they'd hold up against real event maps.
Blurry maps
Non-booth items with numbers or colors
Duplicated numbers
Exhibitor tiers with no booths
Both colors and sizes are used in one tier
Booths already claimed
Size detection isn't always accurate
The edge cases weren't evenly distributed. Each map type gave detection a different level of information, and that changed what organizers needed to do next:
Color/size maps → Detection suggests groupings; organizers bulk-assign.
Numbers-only maps → No grouping; organizers assign one by one.
No detection → Manual setup fallback.
To move quickly, I had Claude to draft the initial flow structure from the requirements and edge cases. It gave me a solid starting point for three-branch flow, but flattened some of the edge cases in ways that wouldn't hold up in practice. I refined the draft, adjusted the fallback logic, and walked the revised version through with the PM before landing on the version below.
USER FLOWS THAT ADAPTS TO WHAT THE DETECTION RETURNS
I also built a matrix mapping every step against every case, putting user story, design requirements, edge cases, and constraints in one place. I brought it to the PM and we worked through the edge cases together, deciding which had to ship at launch and which could wait.
DESIGN SPECS
The core design challenge was making the result review feel effortless. My goal was to build enough trust that organizers didn't need to understand the system. And enough clarity that reviewing AI output didn't feel like more work.
I anchored the design around three principles:
01
Progressive disclosure based on user decisions
Asking users to review all detection output at once would be overwhelming; showing too little would make it hard to make decisions. The design should focus each screen on one decision that moves setup forward.
02
Human-in-the-loop
Even when detection is confident, users should have the final say and the ability to adjust results at key steps.
03
Design for the edges, not just the happy path
Real event maps are unpredictable. Major edge cases surfaced in the audit need a designed fallback that let users continue setup, not an error state that stopped them.
The upload step used to be a passive file dropper. To make it prepare organizers for what came next, I made two changes:
01
Guidance on what details speed up setup
Asking organizers to select what their floormap looks like gave them a chance to learn what works with smart detection.
02
Real-time detection updates
After upload, a progress state showed the system working on the booth detection, so organizers weren't left wondering about the progress.
Once detection finished, organizers needed a moment to review what the AI had done before committing. Two questions shaped that moment:
Where should it live in the setup experience?
How should organizers assign booths to tiers when the detection results give them different starting points?
FLOW EXPLORATION
When detection finishes, users need to check the results before any booths get added. That's a real decision point, so I wanted the layout to give it focus. I judged each option on two things:
Can users tell at a glance which numbers on the map are booths?
Does it hold up for events with lots of booths and groups?
I explored three layouts and walked through them with the internal team. In the end, we picked the modal split view that gave the review its own space without competing with the dashboard for real estate.
LAYOUT EXPLORATION
Maps with booth color/size
Maps with booth numbers only
MAPS WITH BOOTH COLOR/SIZE
Explorations
❌ All booth groups on one screen
Gave organizers a complete overview, but the screen got overwhelming for maps with many groups.
❌ Group-by-group wizard
Focuses each decision to one group at a time, but hard to go back and forth if a group needs editing.
Final design
✅ Collapsible list
Scannable at a glance, focused when expanded, and organizers can jump into any group without walking a sequence.
Maps with booth numbers only
The challenge:
No AI grouping to lean on, organizers had to assign detected booths individually.
01
Tier-by-tier assignment, focus on one tier at a time.
02
Skip any tier that doesn't apply.
Not every event uses every tier. Skip lets organizers move past without forcing an assignment.
Maps with booth color/size
The challenge:
AI could suggest groupings, but organizers needed to catch mistakes before committing.
01
Bulk-assign each detected group to a tier to save time.
02
Flexibility of editing booths in any group before assigning it
With detection and assignment designed case-by-case, the risk was ending up with what felt like four different products under one feature. I audited each path for overlap and consolidated the pieces that could work the same way regardless of how organizers got there.
Before
01
Setup split across three steps: upload, customize tiers, place pins.
02
Editing a pin color or tier description meant navigating back to an earlier step.
After
01
Combining steps 2 and 3 into a single dashboard. All paths (smart booth detection or manual setup) land here.
02
Tiers, pins, and descriptions all editable in one view.
Trade-offs
The dashboard is denser than any single step in the old flow, but it gives organizers a holistic view of their tiers and booths in one place. Every action lives where organizers are already looking, without jumping between steps to edit different info.
