Acelab’s Revit Integration

Acelab data, inside the Revit model

Plugin Design · Systems Design · In-Revit UX

Acelab data, inside the Revit model

Plugin Design · Systems Design · In-Revit UX

The 30-second version

Architects left Revit to use Acelab. Now Acelab lives beside their model.

The problem

Architects re-typed Acelab product data into Revit by hand, on every project.

My role

Lead designer across mapping, keynotes, sync and the panel, working with PM, engineering and CX.

The turn

v1's modal blocked the model, and users asked to start from their selection. We rebuilt around a dockable panel.

Where it landed

A paid Pro feature, used by AECOM, JLL, HOK and Sasaki CONFIRM. The panel is in build.

Team & my role

As the lead product designer, I operated at the intersection of product strategy, user research, interaction design, and technical feasibility

Led Discovery focus groups, CX synthesis, design brief and negotiated design intent

Led research and technical handoff through field shadowing, focus groups, and defensive design specs to ensure data integrity.

Drove build reviews with E-verse wrote objective acceptance criteria, named states, and annotated edge cases.

Architected cross-platform data frameworks and system design to integrate in Revit.

The Acelab-Team + everse Team

Hardik Makrubiya - Lead PM

Armaan Shah - CX

Juliana Zonotto - instructional Designer

Mayank Sharma - BE Dev

Bekele Debisa - FE Dev
E-Verse - Third-Party Revit Devs

Timeline & Status

Timeline Q3 2025 – Q3 2026
Status Live
Premium Paid Feature:
Pro Plan Acelab’s Material Hub

PROCESS

The Problem, the quick TL;DR

Every architecture firm runs two tools, but they’ve never spoken to each other.

Where each decision lives:

Acelab

Acelab is where architecture firms research and manage building products — everything from curtain wall systems to acoustic ceiling tiles.

It’s where material decisions live.

Autodesk Revit

Autodesk’s BIM (Building Information Modeling) software, used by nearly every major firm, where architects model, document, and coordinate.

It’s where the real work lives.

EXISTING STATE

How material mapping works currently?

To design for this complex cross-platform ecosystem, we audited the existing material mapping workflows to understand how architects capture technical data in Autodesk Revit.

Here’s what a normal workflow looked like before the solution existed:

RESEARCH REFRAME

But first, we needed to hear what architects want

We ran contextual inquiry sessions and focus groups with architects and BIM managers at AECOM, JLL, HOK and Sasaki, 3 to 5 people per session, watching how they moved between tools during a project.

8/10

Said they never fully close Revit during an active project

9/10

Said Revit is their real source of truth, not any web tool

10/10

Said they’d use a tool more if it lived inside Revit

HYPOTHESIS

We entered discovery with a hypothesis.

Architects want to push Acelab data into Revit faster.

What we found was different. The workflow wasn’t linear — and it wasn’t one direction.
BIM managers started from Revit. Architects started from Acelab. Neither was wrong. Both were primary. That single observation rewrote the brief.

What they asked for wasn't what they meant

They asked for

They needed

It became our rule

Real-time syncConfidence in the dataSync means trust, not speed

Bulk keynote assignmentLess repetitive workThe real cost is human time

Spec integrationOne authoritative source“Doing the work once”

Rendering software supportWorkflow continuityDon't break their existing tools

Six findings. One rewritten brief, but we still hadn’t answered. Where should Acelab actually live?

1 / 4

Each finding from the focus groups scored by user impact × design impact ÷ effort. Anything above 3.5 shipped in v1.

STRATEGY

Where it lives, and who owns the data

Three paths could connect the tools. Only one put Acelab where the work happens.

CSV import

Creates a new drift problem on every export. It stayed as a way to bring data in, not as the sync.

Web API bridge

Invisible inside Revit, so nobody can see what changed or trust it.

Native plugin

Lives where architects already work, persistent and able to move data both ways.

Rules for who owns which data

Two-way sync can mean two sides overwriting each other, so every field got an owner.

Acelab → RevitAlways

Product name, manufacturer, URL

Revit → AcelabAlways

Family type, count, material, room, area, dimensions

You chooseOne direction, per field, per project

Description, cost, marks, comments, keynotes

Two-way at the product level. One direction per field, so the two tools can't overwrite each other.

Sync modal showing what always syncs and the keynote direction choice

THE DESIGN SOLUTION

Acelab plug-in

for Revit

Acelab plug-in

for Revit

The real insight was that architects never fully leave Revit during a project. It’s their source of truth.

So, We built a native Revit plugin with three core capabilities each one built to eliminate a specific failure in the architects data sync workflow.

DESIGN · V1

First, make the connection work

Two mapping tools, then deeper data with each release.

Map Products

Starts from Acelab. Link a catalog product to the model.

Map Assets

Starts from Revit. Family, type and material in one expandable view, chosen with a user survey.

Dec 25 – Mar 26

Keynotes

Keynote fields and data exchange

Apr – May 26

Six more fields

Description, cost, marks, comments, with direction settings

Jul 26

Shared parameters

Material Hub fields written into Revit

DESIGN · THE TURNING POINT

Version one blocked the workspace

We launched v1 as a modal inside Revit. It mapped products, but it sat on top of the model people were trying to work on, so it didn’t fit how they actually work.

The v1 mapping modal covering the Revit canvas

The modal covers the model you're mapping

What users told us

“

Let me select the family or material first, then map it.

“

The current flow adds too many clicks.

“

Let me search from inside Revit and add it to my project.

200+products is typical. So the real failure wasn't speed. It was a wrong assignment, letting the model and schedule drift apart unnoticed.

Why we moved to a dockable panel

Mapping only makes sense while you can see the model. So the workflow had to stop interrupting it.

‹ ›

V1 · ModalV2 · Dockable panel

Drag to compare the same model

Covers the model

Sits beside the model

Remember the name and search for it

Select it in the model and the panel knows

Open, map, close, repeat

Stays open while you work

Jump to the web app to compare

The same fields and filters, right there

Put it wherever you work

It docks beside the model, floats, or moves to another screen, so the model gets the full canvas.

SECOND SCREEN

Docked Acelab panel with its parts labelled

Follows your selection

No typing the element's name

Shows what's linked

One element, one product

One click to link

Right on the row

Banner when Acelab changes

So nothing updates silently

The search you already know

Filters, sort and fields match the web

Last sync, always visible

Sync stays in your hands

In build

What the panel allowed us to do

From a tool you open to a place you work.

01

Choose with confidence

Narrow 200+ products with the same search, filters and fields as the web, and tell similar products apart in place.

02

Map many elements at once

Select several, see which already have a product, and confirm before anything is replaced.

03

Stay current without leaving

Last sync is always visible, and a banner says when something changed in Acelab.

The bigger unlock

A permanent home inside Revit

The panel connects what you select in the model to your Acelab schedule. New features now have a place to plug in.

Early: press Pick Type

Final: just select

Early idea

Pick mode, or auto-react?

An early concept made you press Pick Type, then click the model. The final design reacts to your selection, so that extra step goes away.

Guardrail

No silent overwrites

Assigning takes one click. Replacing, unlinking or assigning to several elements asks first, and names exactly what will change.

BUILDING IT · DESIGN SYSTEM

The add-in isn’t a web app

Our web design system couldn’t simply come along, so Revit got its own.

The limit

How we worked it out

A Revit add-in is a desktop plugin, not a web page, so web components don't simply carry overA separate Revit UI kit in Figma, designed for a desktop panel

People work in Revit's dark or light themeFigma tokens with light and dark variables, set up with Claude

A narrow docked panel, not a full pageThe schedule table with a pinned Product column, horizontal scroll and resizable columns

Users already know the web scheduleThe same search, filters, sort and fields, so nothing needs relearning

BUILDING IT · DECISIONS

Four calls, and what each one cost

Each one had a real alternative. Here’s what we picked, what we rejected and what we gave up.

Both ways on every field

One way, set per project

01

One direction per field

Two sides writing to the same field means one silently overwrites the other. A per-field setting keeps it predictable.

What it cost: A field can't be edited from both tools in the same project.

Auto-sync on open

Manual, with a banner

02

Sync stays a button

Real-time was the top request, but automatic sync is on hold. A banner says when Acelab changed, and the person decides when it reaches the model. CONFIRM

What it cost: People have to sync to see updates, so the panel isn't live.

Assign both ways in the panel

Revit → Acelab only

03

Panel adds no new sync logic

The existing sync engine handles everything else. The only new sync surface is the update banner, so nothing that already worked is put at risk.

What it cost: You can't push an assignment from Acelab through the panel.

Panel reacts to tags, plus a tag manager

One family, one click to place

04

Keep Product Tags small

A placed tag is an ordinary Revit object, configured once per project. That solves the drawing problem without a new system to learn.

What it cost: Clicking a tag doesn't update the panel, and types are managed in the Project Browser.

BUILDING IT · ROADBLOCKS

What got in the way

The hard parts weren’t on the screen. Each one shaped the design.

01ResolvedCONFIRM

Two-way sync raised model bloat and performance worries

Syncing everything both ways put the model itself at risk.

A short, defined field list, with one direction per field. Everything else stayed out of scope.

02Designed

Revit can't write to protected folders

Keynote files saved in the wrong place simply failed.

Named the real cause in plain language and sent people to choose another location.

03Designed

Firms run different plugin versions

Staying current has to work around real project deadlines.

An update prompt with “Update on Exit”, a clean uninstall flow, and supported versions tracked as a guardrail.

04Designed

Failures left no trail

When something broke, we needed to see why.

A pre-filled error report that sends recent logs and never includes model geometry.

Every state designed, not just the happy path

Holds the layout while data arrives

No match

Points to the next thing to try

Couldn't load

Says what happened, with a retry

WHAT’S NEXT

Three problems the panel opened up

Each one came from a real customer request.

Product TagsIn buildInterior designers draft instead of model, so nothing could be linked. One click puts a product on the drawing, and removed products flag their tags, never delete them.

Sync SummaryIn researchCustomers asked for a change log. It shows old and new values, and won't say “complete” if the upload failed.

Multi-model linkingOn holdOne project, many models. We mapped every scenario against every field type, like a second model double-counting a row.

IMPACT

Where it stands

PRO

A paid feature, gated by subscription

projects in the initiative, 3 complete and more in progress

firms using it: AECOM, JLL, HOK, SasakiCONFIRM

How we'll measure it, tracking is still being built

Driver

Install → first successful sync

North star

Weekly linked projects with a successful sync

Guardrail

Sync failures, users on a supported version

TAKEAWAYS

What this project taught me

01Trust beats speedPeople asked for real-time. They needed to see what changed.

02Ship the connection, then listenv1 made it work. Usage showed us how it should work.

03Decide who owns the dataThe hardest design work here was never on a screen.

04Design for the hostRevit sets the rules. The kit, tokens and states had to fit inside them.

Other Projects

All Projects

Data-request loop for architects

RFI Loop

Design system · Accessibility

Covesive Design System

AI-powered design tool

Ada Tech x PUMA

🌱

Work

About

Lab

Resume

Acelab’s Revit Integration

Acelab’s Revit Integration

Acelab’s Revit Integration

Every architecture firm runs two tools, but they’ve never spoken to each other.

Where each decision lives:

How material mapping works currently?

How material mapping works currently?

RFI Loop

Covesive Design System

Ada Tech x PUMA

Beyond Work or
know more About Me! :)