How Would You Build Bloomberg for Property?

Bloomberg won by putting the market and the means to act on it on the same screen. Property has the same gap — and most of the field is standing on the wrong side of it.

At Giraffe we often talk about building the Bloomberg for property, and for us it was never just about data.

Bloomberg's edge wasn't the feed — plenty of other firms sold market data. Bloomberg won by putting the data and the analytics on the same terminal: the one screen where you both see the market and act on it. The value was in the union. Property has the same gap, and most of the field is standing on the wrong side of it.

A site is not worth what it is

You can know everything static about a parcel — every line, every control, every setback — and still not know what it's worth. A site isn't worth what it is today; it's worth what it could become. Value is set by the highest and best use, and site attributes don't tell you what that use is.

The property-data industry lives in the physical and legal — geospatial, zoning, exactly the layers everyone is racing to assemble. The vendors stop precisely where value begins, because what's financially feasible and maximally productive aren't data. They're a design problem and a financial problem, and you can't look them up. You generate them.

The balance sheet and the P&L of dirt

You can't value a company off its balance sheet. It's a snapshot of what exists, which is why book value rarely equals market value — it says nothing about what the assets earn. For that you need the P&L. A site is identical. The site data is the balance sheet: the static inventory of what you've got and the constraints you work inside. The feasibility — the scheme you can design and the pro forma it throws off — is the P&L: the forward-looking flow, the productive capacity of the dirt under its best use.

Two sites with identical data — same area, same zoning, same controls — routinely trade at very different prices. Same balance sheet, different P&L, different value.

Price a site off the balance sheet alone and you've repeated the book-value mistake. The proof is in every market: two sites with identical data — same area, same zoning, same controls — routinely trade at very different prices. Same balance sheet, different P&L, different value. The entire gap is what someone can design and finance there, and no data layer can see it.

Why the two halves never met

Both halves are obviously required, yet nobody puts them together, because they've always lived in different tools. The balance sheet sits in GIS and planning portals. The P&L gets built later, by different people, in CAD and spreadsheets. Every valuation means reconciling the two by hand — re-keying constraints, re-drawing schemes when the data changes, re-running numbers. That reconciliation is pure integration labour: the tax you pay for keeping an asset's two statements in separate buildings.

The terminal

Giraffe is built around this. Spaces is the balance sheet — filter, search, cut views of your site data to find opportunities and manage the pipeline. Design is the P&L — generative algorithms and deep analysis to determine the highest and best use. The same geometry underlies both, so the numbers move when the scheme moves, and the scheme updates when the data does. No re-keying, no reconciliation by hand.

A data layer tells you what a site is. A feasibility tells you what it's worth. You need both on the same screen, or you're pricing an asset off half its accounts.