ABI - BID - Revit, Experience Design
Revit Wall Modeling Projects
- Create Wall By Segment/Room
- Create Hosted Wall On Host Wall
- Explode Wall Into Multiple Walls
Revolutionizing Wall Modeling in Revit:
Leading a Multi-Feature Initiative to Modernize a Core Workflow for Millions
My Role:
Lead Experience Designer & Researcher. I owned the user research, discovery, and UX design for this entire initiative, from initial concept validation through to detailed feature specification and customer validation.
Key Contributions:
Generative & Evaluative Research, Design Strategy, UX Design, Customer Community Building, Cross-Functional Leadership.
The Challenge
For over a decade, architects using Revit faced a tedious and error-prone process for modeling multi-layered wall assemblies (e.g., structural core + interior finish + exterior cladding). This "core and finish separate" workflow, used by 68% of architects, involved manually creating, aligning, and managing multiple wall elements. This was slow, inflexible, and created a maintenance nightmare for complex buildings.
The Vision
To transform this manual chore into an intelligent, automated, and data-rich process through a suite of interconnected features: Create by Room/Segment, Hosted Walls, and Wall Explosion.
Before:
Users can only use the Draw Gallery Tools to create walls, which leads to repetitive clicks and time waste.
After:
Users can create multiple walls with just one click and can easily define the wall relationships which meet the construction reality.
01
The Proof is in the Data: Understanding the User
68% of Architects Were Struggling
Our initial survey of 225 architects and BIM managers confirmed the scale of the problem. The vast majority were using inefficient manual methods, costing them hundreds of hours per project.
Why users don’t use existing Revit Wall feature? What’s the Workaround?
From the research, we’ve tried to understand why users hate the existing Revit Multi-Layer Wall Type, and what are their workarounds.
Key Research Insights
- Inefficiency: Users wasted time manually drawing, aligning, and joining walls.
- Fragility: Changes to the core structure meant manually updating all attached finish walls, a error-prone process.
- Missing Data: Revit had no native way to understand the logical relationship between a structural core and its finishes, limiting automation and data extraction.
- High LOD Hurdle: Moving from a simple (LOD 100) design model to a detailed (LOD 300-400) construction model was a brutal, manual process of breaking apart walls.
Prioritization
After the research, we’ve worked on brainstorming new feature ideas to solve users pain points. And I conducted another round of survey to prioritize these features. Thus, I helped team defined team’s focus for the next release.
02
A Three-Pillar Solution
A Phased, Cohesive Strategy
Instead of isolated features, we designed a synergistic suite that addressed the user's journey from creation to modification to detailing.
Pillar 1: Create by Room & Segment (R2026)
The "Creation" Boost. Allows users to generate finish walls with incredible speed by simply clicking a room boundary or a wall segment. This solved the initial modeling inefficiency.
Pillar 2: Hosted Walls (R2027)
The "Intelligence" Layer. Adds a logical relationship between walls. When the host (core) wall moves or rotates, its hosted (finish) walls follow automatically. This solves the maintenance and change management nightmare.
Pillar 3: Explode Wall (Future)
The "Detailing" Power Tool. Allows users to take a simple, generic wall and intelligently break it into its constituent parts (e.g., core, sheathing, finish) for detailed construction modeling, solving the high-LOD transition problem.
03
Deep Dive: The "Create by Room & Segment" Feature
What is it?
In the easiest way, the two features allow users to create walls on existing walls by a click.
The Design Challenge
How to make segment selection intuitive and predictable, especially around complex elements like columns? Examples are as following:
Design Iteration & Validation
To solve these design problems, I’ve kept iterating the design and validate with users in each Sprint Demo, and in user community like Beta Forum.
Key Design Decisions
- Tab Key Toggle: We used the familiar Tab key to let users toggle between selecting the segment between columns or the entire wall length.
- Room Bounding Logic: We initially used room-bounding rules but pivoted based on user feedback that it was too restrictive.
- Integrated UI: We embedded the new tools directly into the existing wall creation ribbon to avoid context switching.
From 51 Seconds to 31 Seconds: Quantifying Impact
The "Create by Room & Segment" feature was our first deliverable, focused on pure efficiency gains. We meticulously measured the old workflow and designed the new one to slash time and clicks.
04
Deep Dive: The "Hosted Walls" Feature
Beyond Movement: Designing a New Relationship Model
While Create wall by segment/room was to help users boost modeling efficiency, Hosted Walls was a paradigm shift. It wasn't just about graphics; it was about teaching Revit what it meant for one wall to belong to another.
The Research Challenge
Users had deeply nuanced expectations about this relationship. Should finishes follow a host wall if its height changes? What if it's tapered? We conducted over 10 interview sessions to map these mental models.
Key Design Decisions
- Follow Existence & Position, Not Shape: We designed the system so hosted walls follow if the host is moved, rotated, or deleted, but not if its profile or height is edited. This matched user expectations for flexibility.
- Clear Visual Language: We added subtle visual indicators in the UI to show the host-hosted relationship.
- Robust Error Handling: We designed clear messages for edge cases (e.g., "This wall can no longer host its dependent walls because they no longer touch").
05
My Discovery Engine: Building Trust & Getting Feedback Fast
Building a Continuous Feedback Loop
Traditional, periodic research wasn't enough for a project of this scope and speed. I pioneered new methods to get high-quality feedback continuously.
1. The Focus Group Teams Channel:
I created a dedicated Microsoft Teams channel with 30+ passionate users. This became our rapid-fire testing ground.
Result: 90+ posts, resolving 10+ design problems. We could get feedback on an icon design in hours, not weeks.
This method has been chosen to be presented at the largest internal event TechX.
2. Fast discovery
I create a first iteration loop with different methods to help boost the speed of discovery. With this method, our team can complete two discoveries within two weeks.
This method has also been chosen to be presented at design meet-up sessions.
3. Inside-the-Factory (ITF) Events:
I led hands-on testing sessions at Autodesk's ITF events, where we gathered deep, qualitative feedback on working prototypes directly from our most engaged users.
This multi-pronged approach ensured we were never building in a vacuum.
06