Context Is King. But the Industry Is Solving Only Half of It.

A First-Timer’s Reflection From Realcomm 2026

Three days of AI sessions at Realcomm 2026 kept returning to one word: context. The industry is investing heavily in internal context – making your own data legible to AI – and that work matters. But it’s only half the story. External context (what’s happening across the market, the sector, and the cycle) matters just as much. Internal context tells you what your number means; external data context in commercial real estate tells you what your number means in the market, and that’s the half that turns a correct number into a correct decision.

The AI demos were polished. The sessions were packed. The energy was real. But the takeaway that holds up after three days sits underneath all of it – one word: context.

It ran through the data architecture sessions, where teams described the meticulous work of adding context to their own data legible to AI – the definitions, the lineage, the metadata that tell a model whether “NOI” in one system matches “NOI” in another. One panel put it plainly: context is king. Another offered the line we keep coming back to – “that’s not a hallucination, that’s just bad data.” 

It’s important work. But after three days of listening, we couldn’t shake a quieter observation: the industry is solving only half of the context problem. 

What Internal Context Is, and Why Everyone’s Working on It

The half getting the attention is internal context, making your own data legible to AI. That’s necessary. Without it, a model will confidently average two things that were never the same. But internal context has a ceiling. It can tell you what your number means. It can’t tell you what your number means in the market. 

Why Clean Data Isn’t Enough: Transparency Is Not Interpretability

A General Partner on one panel said something that stuck with us: transparency is not the same as interpretability. He described an asset reporting 27% occupancy – perfectly accurate, perfectly clean, and perfectly misleading, until you know there was a fire and the building is being re-leased. The number wasn’t wrong. It was missing its context. And no amount of internal data hygiene would have surfaced that, because the context lived outside the dataset. 

That’s half of the problem we heard far less about: external context. What’s happening across the market, the sector, the cycle. What the public markets are signaling about private pricing. The context that turns a correct number into a correct decision. 

Green Street: Where External Context Has Always Come First

This is the gap Green Street has spent forty years filling – it is, frankly, the problem our research was built to solve. Green Street’s work has always been about the context around the number: independent research, real-time news across the real assets markets, and the public-to-private signal that has pointed to where private capital moves.

In an AI world, that kind of context becomes more valuable, not less. Models are very good at producing fluent answers – and have no idea whether to believe them. External context is what tells you whether to. 

The Stage, and Everyone Not on It

Realcomm’s stage belonged to the enterprise: the largest institutions, the global consultancies, the proptech platforms. Their reality is a long, costly build: resolving data, layering context, re-engineering the organization around it. And while the technology team builds, the business still has to operate. Which leaves a real question: what’s the interim play while that build is underway? Competition doesn’t wait for a finished data model. And context isn’t a project anyone finishes – first-party data evolves and markets move, so it has to stay live and it has to be trusted. Context pulled from the open web is just a faster way to be confidently wrong. 

What About the Mid-Market? 

There was a quieter question the stage didn’t dwell on: what about the mid–market? They live a different reality: leaner teams, tighter budgets, none of the scale. And also, none of the drag: no decade of accumulated data debt, no sprawling stack to untangle, no eighteen-month readiness program standing between them and action. 

But the private conversations told a different story, and the mid-market is telling it. They see momentum like never before. More and more workflows are getting augmented; every team member is leveling up. And the savvy experts get dangerously better – they don’t offload their thinking to AI – they amplify their judgment with it, grounded in trusted context that propels them not just toward faster decisions, but toward action. 

What We Took Away

If Realcomm left us with one conviction, it’s this: the edge is the context that tells you whether to trust the answer – and what to do next. That data context in commercial real estate has always been as much external as internal. The teams that take this in will make better calls, and faster – while others are still cleaning their own data. 

The energy fades. The context stays. And we know which half of the problem matters most.

Frequently Asked Questions