Work Transformers Labs

We research where the built environment is going.

And we turn what works into practice: the advice and frameworks our clients rely on.

Applied researchIndustry intelligenceFuture operational systems
Why we built Work Transformers Labs

We kept seeing the same problem across the industry. Teams were overwhelmed by documents, disconnected systems and increasing pressure to move faster. Labs exists to explore how AI can improve real operational work across the built environment.

Labs

Inside the lab, where the research is built.

Systems alone aren't enough. What makes them work is the reasoning behind them.

Latest researchbecomes logicInputsReasoningDecisionbare system: skips reasoning, stalls
Research becomes the logic that powers reasoning. Without it, a system can move documents but cannot judge them. This is where bare systems fail.
How the Labs works
We research

The pressures and openings reshaping property, projects and capital.

We test

On real work, with real clients, not in the abstract.

We turn it into practice

What works becomes advice and frameworks we deploy. What doesn't sharpens our thinking.

What the research surfaces
From the research
EC2 prime office yields hold at 6.1% in Q2 2026Destination
Fit-out cost overruns widening to +12% on recent projectsLighthouse
Covenant risk rising across secondary retailGhostmap®
Hybrid is stabilising: three days the emerging office normDestination
MEES deadlines pulling refurbishment decisions forwardLighthouse
Where AI helps a review, and where judgement must stay in chargeHorizon
A sample of what our research surfaces.
From research to deployment
Applied research, deployed and kept sharp.

Labs is the engine behind the systems we deploy. Research does not sit on a shelf; it runs in production, and what production teaches feeds back into the research.

01
Applied research

We study where the built environment and AI are heading, and turn it into method.

02
Deployed into live systems

The method goes into systems running inside client firms, not slideware.

03
Production signals captured

Real usage shows what works, what drifts and what to improve.

04
Research sharpened

Findings feed the next iteration, so deployed systems keep getting better.

A closed loop: research informs deployment, deployment informs research.

Deployed learnings
What Labs has learned from real systems.
System intelligence

Reviewers move to exception-based checking once the system shows its confidence and sources.

Operational patterns

The same five-stage shape recurs across review, underwrite and reporting work.

Benchmarking intelligence

Deviation flags earn trust when every one is tied to a named source.

System evolution

Teams adopt faster when the system explains its reasoning, not just its answer.

Production-informed systems

Escalation thresholds are tuned from real client decisions, not assumptions.

Enterprise feedback loops

Each deployment improves the shared method, without sharing any client's data.

Our research programmes

Six long-running programmes. Each studies a different part of how the built environment is changing, and feeds back into the work we do with clients.

/ 01
Project Destination
Ongoing programme

Workplaces people actively choose to use, not just occupy. What makes an office worth the commute, and how organisations design for it. Hybrid made this the real test of any workplace.

Includes the Destination 2.0® framework, a research programme and book, with benchmarking and industry roundtables.

What we are learningWhat separates the places people return to from the ones they avoid.
Read this research →
/ 02
Project Ghostmap
Ongoing programme

How investment and real estate teams underwrite and assess deals faster, without losing rigour, so a lean team can look at more of what matters.

A study of the reasoning behind better, faster investment decisions, and how it can be made consistent and traceable.

What we are learningWhere speed and rigour reinforce each other, rather than trade off.
Read this research →
/ 03
Project Horizon
Ongoing programme

Where AI genuinely helps an expert team, and where human judgement has to stay in charge. Most firms are adopting AI with no clear answer to that question.

What we are learningThe dividing line between work to delegate and decisions to keep.
Read this research →
/ 04
Project Lighthouse
Ongoing programme

How organisations spot operational risk and change before it costs them, by keeping decisions under review as the world moves, not only at the moment they are made.

What we are learningHow to see change early enough to act, not just record it.
Read this research →
/ 05
Project Aperture
Ongoing programme

How an operational system should present a complex decision: composing the right view for each task instead of forcing the work into a fixed dashboard.

From static components to generative interfaces, and what stays fixed so people keep their bearings and their trust.

What we are learningWhat an interface that fits the decision, not the other way round, actually looks like.
Read this research →
/ 06
Project Thunderpoint™
Ongoing programme

The harness around AI: the guardrails, verification, context control and human checkpoints that make a system dependable for high-stakes work.

Why reliability is engineered around the model, not prompted into it, and how silent failures are caught before they matter.

What we are learningWhat it takes to make AI output a serious firm can stand behind.
Read this research →
Events and webinars
Where the research meets the room.

We convene senior leaders across corporate real estate, law, technology and investment, and take the research to the industry stages.

See all events →
How we publish

What we share, and what we keep.

Our programmes are open at the level of questions and principles: what we are studying, what recurs across real deployments, and what holds up under pressure. The methods, the client data, and the parts still being worked out stay with us. We share what helps the field think clearly. We keep what is still being earned.

Work Transformers Labs

Want the research as it is published?

We share what we are learning as it develops. No noise, just the work.