About Speckle
The data intelligence layer for AEC.
Every project generates thousands of hours of design and model data. Speckle turns that untapped insight into an enterprise asset that is accessible, reliable, and ready to drive decisions.
Design data in
Ask, compare, validate and prepare model data
Tower A · Models
Structure
Architecture 2 issues
Facade
Coordination
Deliverables out
Generate project intelligence, reporting and handover outputs
Tower A · Cost & carbon report
Live · v24Steel
1,284 t
−3.2% vs v23
Carbon
412 kg/m²
−5.1% vs v23
Model health
96 %
+4 pts
Concrete by level
With Speckle, every project informs the next. Historical data becomes portfolio intelligence, delivering sharper bids, stronger benchmarks, and more informed delivery decisions across your entire portfolio.
The challenge
Your firm doesn’t have a data shortage, it has a trust shortage.
Across hundreds of models and projects, teams lose 28–46 hours a week per project to manual validation, QA/QC, data conditioning, and reporting. Project standards vary by project, partner, and platform, meaning that cross-project analysis requires time-consuming data wrangling and manual reconciliation before anyone trusts the results.
Workflow friction
Critical VDC & BIM delivery work still depends on exports, cleanup, and re-entry across disconnected tools and teams.
Lack of trust
Model data standards vary by project and partner, while inconsistent enforcement leaves models too unreliable to build predictive or generative AI on top of.
The human impact
Thousands of expert hours are wasted on low-value data work, slowing decisions and making it difficult to scale capacity.
Source for 61.1% and 5–10%: Speckle, The State of Data Readiness in AEC (2026).
Open source since 2016 Dimitrie Stefanescu
Founder & CEO
Why Speckle
Built in the open, by people who were tired of rebuilding the same pipelines.
Speckle started as an open-source project in 2016 because the AEC industry’s data problem wasn’t a missing feature, it was a missing layer. Every firm was solving the same integration, validation, and reporting problems alone, inside whatever tools they happened to work in. And no incumbent vendor was incentivized to fix that because their business models depend on the very friction a shared layer would eliminate. So we built Speckle to be that layer: open at the core, platform-agnostic, and designed to sit across the tools your teams already use to make them more effective.
That’s still true today. Speckle isn’t another CDE, BIM authoring tool, or BIM collaboration tool. It’s the connective, governed intelligence layer that makes every other tool in your stack more valuable, because the data flowing through it is finally something you can trust.
We work with the firms on the edge of AI and data innovation, who choose to leave the data plumbing to us so they can focus on the value-add workflows that build competitive advantage that compounds over time. That’s what makes them the most profitable, most successful designers, builders, and operators in the industry.
How Speckle works
From raw model data to trusted decisions in three steps
01 · Access
Connect & normalize
Speckle connects model data from every design tool, file format, and project team into one normalized layer without asking anyone to change how they work.
- AEC data integration layer Reduce integration cost
- Enterprise data lake integration Enable your broader data strategy
02 · Governance
Validate & control
Automated checks catch data errors and omissions before they hit coordination, estimating, or reporting. Data conditioning makes it easy to scalably apply consistent naming, parameters, and classification standards.
- Data validation & QA/QC Reduce risk, protect margins
- Data conditioningCreate reliable inputs for AI and analytics
03 · Act
Put data to work
Once data is accessible and trusted, it becomes usable, powering real-time project intelligence, portfolio benchmarking, and reporting that doesn’t get rebuilt from scratch every time.
- Project intelligence & reporting Win more work
- Cross-project benchmarking Build portfolio intelligence
Speckle turns fragmented model data into a governed enterprise layer for faster decisions, less manual effort, and AI-ready intelligence.
How we’re built
No lock-in. No single vendor ecosystem. We deliver the building blocks, your data and intelligence stays yours.
Speckle’s core is open source, meaning the layer sitting under your most critical project data is transparent, extensible, and open. And because Speckle is vendor-agnostic by design, it connects to the tools your teams already use — from authoring tools to analytics platforms — instead of asking your firm to standardize on one ecosystem.
View the source on GitHubWe’re no longer solving data problems one project at a time. We’re building a repeatable data layer that can support multiple tools, multiple offices, and multiple projects.
Matt Lewis
Associate Director
Trusted by leading AEC firms
Enterprise teams are already compounding their competitive advantage by building on Speckle.
Speckle makes our design data accessible across construction, helping us process it and maintain clean, consistent business logic.
David Morgan
BIM Data Engineer
It’s how modern digital tools should operate — it connects to your data, automates your workflows, and doesn’t lock your information away.
Emiliano Lupo
Design Technology
You’ve already got the data. Let’s put it to work together.
Whether you’re evaluating a first workflow or planning a portfolio-wide rollout, we’ll show you what Access, Governance, and Transformation look like on your own project data.
Our investors
Backed by investors who believe in the power of open data
We’ve raised from investors who care about long-term impact, not hype. Our backers are committed to the same thing we are: a shared data layer the industry can build on for decades, not quarters.
