Vantrow
Book an intro call
← Insights

Article · 2 min read

Why 92% of CRE firms pilot AI — and so few reach their goals

Andrew Brown · Jul 13, 2026

Quick answer

CRE firms pilot AI almost universally — Colliers puts it near 92% — but only a small fraction reach their goals, because pilots run beside the workflow instead of inside the system of record. The firms that get value fix the record first, stage AI output for human approval, pick one weekly workflow, and give it an operational owner.

The adoption numbers and the outcome numbers tell two different stories. Colliers reports that roughly 92% of CRE firms have piloted AI while only about 5% say they've achieved their AI goals — a spread Colliers calls the AI productivity gap. PwC and ULI's Emerging Trends in Real Estate finds the same shape: AI a stated strategic priority at over 90% of leading firms, active pilots at well over half, embedded results far rarer.

Why do CRE AI pilots stall?

Because most pilots are beside the workflow, not in it:

  • The pilot summarizes documents into… another document, which someone re-keys into the tracker.
  • The output has no review step, so nobody trusts it enough to act on.
  • The pilot lives with one champion; when their attention moves, it dies.

None of these are model failures. They're integration and governance failures.

What separates the 5% that get value?

The pattern across the firms that graduate from pilots is consistent:

  1. A system of record exists first. AI output lands as structured data in the operating system, not prose in a memo.
  2. Governed autonomy. Suggestions are staged for human approval — which is precisely what makes teams comfortable using them. (Propose, never commit.)
  3. A narrow, daily wedge. One workflow that runs every week — broker updates, lease abstraction, deal capture — instead of a moonshot.
  4. An owner in operations, not in innovation.

Is "agentic AI" going to change this?

The industry expects autonomous multi-step agents to reach mainstream use in 2026–27, and they will genuinely widen what's automatable. But agents raise the stakes on exactly the thing pilots already lack: governance. An agent that can execute a workflow can execute a mistake at workflow speed. The bridge across the productivity gap isn't a stronger model — it's the boring machinery of staging, review, and audit that lets a firm say yes to automation without handing over the keys.

What should a mid-size firm actually do this quarter?

  • Pick the one workflow the team does weekly and hates (broker updates is a common winner).
  • Make sure its data lands somewhere structured — fix the record before the robot.
  • Require review-before-commit on anything AI writes.
  • Measure one number (hours to process, days to follow up) before and after.

That's how a pilot becomes an operation — and how you end up in the 5%.

FAQ

What is the AI productivity gap in commercial real estate?
Colliers' term for the spread between near-universal AI piloting (~92% of CRE firms) and rare goal achievement (~5%). Firms experiment widely but struggle to embed AI in core workflows, so pilot learnings never become operating results.
Why do CRE AI pilots fail?
Three repeat causes: output lands beside the workflow (a summary document instead of structured data in the system of record); no review step, so the team never trusts it; and champion-dependence, where the pilot dies when one person's attention moves. All are integration and governance failures, not model failures.
What should a CRE firm automate first with AI?
The narrow workflow the team runs weekly and resents: processing broker updates, abstracting leases, or capturing deal-flow notes. A daily-use wedge with structured output and human review builds the trust and the data foundation that bigger automation needs.
Is agentic AI ready for CRE operations?
It's arriving — industry analyses expect agentic systems to hit mainstream use in 2026–27 — but agents amplify the governance requirement rather than removing it. An agent executing a multi-step workflow can execute a mistake at the same speed. Staging, approval, and audit remain the bridge.

See what this looks like for your firm.

Governed software, configured to how you actually work — built embedded, shipped as something you own and can audit.

Book an intro callSee what we build

Keep reading

Article

The Startup "Fitness Stages" Lesson, Read for a Growing CRE Shop

A construction-fintech CEO's "5 fitness stages" framework was written for founders — but it answers a question every growing CRE shop is already asking: as headcount and deal volume climb, what do I let my systems do on their own, and what still needs a human hand on it? We translate startup growth stages into a staging discipline for the tools an owner-operator buys.

Article

A $3.5B Valuation Is an Operations Story, Not a Building Story

Brookfield is buying into the $3.5 billion Hudson Square portfolio as tech tenants push rents up roughly 20%. But a leasing surge rewards the owner whose process can absorb it — capturing tours, broker updates, and lease expirations in one place — not the one with the best building. Here's what to fix before demand tests your pipeline.

Article

Undervalued CRE Sectors: The Edge Isn't the Tip, It's Your Data

Everyone wants the overlooked CRE sector — IOS, small-bay flex, secondary medical office, tertiary self-storage. But the sector is only half the answer. The durable edge is being organized enough to source, underwrite, and close a niche deal before the big funds crowd in. That edge comes from a clean system of record, not a hot tip. Here's what operators keep naming, and what actually lets you move on it first.