AI in commercial property management: what CBRE, JLL and Cushman & Wakefield are doing now - and what's next
AI is moving from slideware to line-item results in commercial real estate. Leaders at CBRE, JLL and Cushman & Wakefield are applying it to leasing, facilities services, contract management and research - with a clear throughline: proprietary data is the moat, and efficiency is the first dividend.
The message is consistent across recent calls and briefings: get disciplined about ROI, protect and productize your data, and equip people - not replace them. Here's how the big three are executing, and what it means for owners, occupiers and FM leaders.
Cushman & Wakefield: Reading the signals and wiring the enterprise
Cushman & Wakefield launched an AI Impact Barometer to model how AI reshapes sectors and asset classes. Early signals point to pressure on white-collar roles exposed to automation and rising vacancy risk in Class B and C offices as back-office demand thins. Leadership has been blunt: AI will create winners and losers.
On the ground, they're using AI to generate insights inside their asset services platform and to accelerate space planning in global occupier services. In leasing, tools like OneAdvise assemble digital tour books, benchmark deals and support negotiations - feeding a larger data lake that powers cross-sell and client intelligence.
Beyond core ops, AI is tracking cross-selling opportunities, refining compensation, strengthening CRM in capital markets, and monitoring legal obligations in contracts. The aim is to break organizational silos by making data flow freely, not just redrawing org charts.
CBRE: ROI-led efficiency and turning a vast data estate into an edge
CBRE frames AI in two lanes. First, deploy it where it beats traditional efficiency levers like offshoring - with disciplined trade-off analysis before investing. Second, convert what it calls the industry's most comprehensive real estate data into a durable advantage.
Leadership groups risk and opportunity into three buckets: transactional businesses (brokerage), investment activities (creating and improving physical assets), and property/facilities management. Brokerage and investment are comparatively insulated because clients pay for creativity, strategy, negotiation and relationships - areas AI supports but does not replace in the near term.
Facilities and property management sit closer to AI disruption as data-heavy workflows get automated. That brings upside and risk: AI can enable or disintermediate knowledge flows. CBRE expects by the end of 2026 to show tangible gains from extracting, assimilating and delivering its data to professionals, cutting data acquisition and research costs while making brokers more effective.
JLL: Efficiency first, built on a decade of data work
JLL credits AI with helping margins over the last two years, growing revenue without adding headcount. Tools surface broker opportunities and give property management teams live pricing and execution guidance for work orders - with impact across workplace management as well.
A long-running push to standardize and own data underpins this. Through JLL Spark and internal platform investment, the company centralized proprietary datasets across countries. The stance now is clear: build in-house, protect the data, and avoid overreliance on third-party proptech when it risks giving up the "secret sauce."
JLL also stresses a human-led model. The complexity of assets, local expertise and fiduciary duties set hard limits on replacing professionals. The edge comes from people equipped with better data - which also lifts revenue per head across transactional and service lines.
What this means for owners, occupiers and FM leaders
AI is already paying for itself where workflows are structured, repetitive and data-rich. It's also exposing a demand gap in legacy office product that depends on back-office roles. Meanwhile, client-facing advisory and investment work remains human-centered, with AI augmenting rather than substituting.
- Near-term wins: lease abstraction and comp benchmarking, digital tour books and proposals, FM work order pricing and triage, technician assist, and contract obligation tracking.
- Enterprise value: cross-sell mapping from a unified data lake, broker enablement with faster insights, and lower research/data acquisition costs.
- Risk lens: Facilities and property ops are most automatable; brokerage and investment depend more on human judgment and relationships.
A practical AI playbook for 2026
- Build the data backbone: Inventory data sources, standardize IDs (assets, leases, vendors), define ownership/governance, and set access controls for sensitive fields.
- Set two investment theses: Efficiency (benchmark against offshoring/BPO, target cycle-time and unit-cost cuts) and differentiation (turn proprietary datasets into tools and insights clients can't get elsewhere).
- Prioritize FM use cases: Predictive maintenance and anomaly detection, dynamic vendor/pricing guidance, and field technician copilots. Track OPEX per square foot, first-time fix rate, and SLA adherence.
- Broker enablement: Opportunity scoring by client/sector, auto-assembly of tour books and pitch content, comp-set QA and scenario modeling. Measure revenue per head and proposal win rates.
- Legal and finance: Extract obligations, options and triggers from contracts with alerts; model commissions and rebates; tune compensation plans with transparent logic.
- Change management: Train teams, codify "human-in-the-loop" checkpoints, create prompt/playbook libraries, and establish a simple RACI for AI-assisted decisions.
- Risk and compliance: Model risk management, data residency controls, IP/privacy guardrails, audit logs, and third-party vendor governance.
- Value tracking: Baseline before deployment. Report monthly on cycle times, margin impact, revenue per head, research/data costs, time-to-lease, and vacancy forecasts by asset class.
The moat is proprietary data
The common thread across CBRE, JLL and Cushman & Wakefield is simple: scale and cleanliness of first-party data decide who pulls ahead. Efficiency pays the bills; unique datasets win the mandate. If you build with vendors, keep rights to your derivatives, model outputs and feedback loops.
For deeper how-tos and playbooks, see AI for Real Estate & Construction and AI for Executives & Strategy.
AI won't replace the people who own the client relationship, read a market and get a deal done. It will reward the firms that turn their data estate into margin, speed and better outcomes - consistently.
Your membership also unlocks:
