What CRE Really Wants From Data

Cherre, with its “real estate data management and intelligence” platform, has just raised a $30 million series C round, led by HighSage Ventures and including Nuveen Real Estate, RXR (RADV), certain principals of TA Realty, and others. Plus current investors, Trustbridge Partners, Glilot Capital Partners, Intel Capital, and Carthona Capital helped the cause.

At this point, the typical continuation would be what Cherre plans to do, which is “invest in expanding its data intelligence and insight capabilities” and provide clients tools to do more and better.

Every company has an angle, and Cherre, like all, takes a particular direction. CEO and co-founder L.D. Salmanson spoke with GlobeSt.com to discuss some of the complications of data and the real intelligence behind decisions, which is ultimately the user companies.

He said the product works well now, but there are areas where they want to use their engineering staff to expand because it’s what customers need. Like anomaly detection.

“It could be as simple as we expect this financial data looks like this and it isn’t, or the data is stale or there are rows missing,’” Salmanson told GlobeSt.com. It may sound like trivial concerns, but if you’ve ever had to clean data, find problems in formatting, or wonder whether what you think you see is what someone put into a database, it immediately gets twisted and painful. It also never sees a final resolution because there is always new and unique data.

A classic problem in corporate data, for instance, is that different parts of a company use the same term for what is inherently different data. Say that a large property owner and operator has to track consumables. Inventory cost could be a first-in, last-out model or an ongoing updated average price or first-in, first-out. Any of these could be valid, but different groups might use different definitions without realizing it. The pricing data could come in from a vendor who uses other price calculations. Or the vendor might round numbers differently from the business, with small discrepancies eventually building to significant sums.

Another direction Cherre plans is more data quality assurance with incorporated business rules that could help identify how the information should be used and incorporated. “A lot of products look really good,” Salmanson says. “Maybe their job is to provide a point solution. But the data behind the scenes [could be] really bad.” And the CRE companies need data that can help them see an ROI as quickly as possible — and they need to understand what the data is and how any manipulations work.

“Our clients are really smart,” he said. “They have sophisticated engineering teams. Get us off the ground and give us the infrastructure and tools that let us build more and more things, but we’re not looking for black box AI or magic.”

Ultimately, what Cherre wants is a world of data that can be connected in practical ways. “Just ask the questions,” Salmanson says. “Real estate operators are not stupid; they know what they want. If data is connected, I can ask questions. If it’s not, nothing can be automated.” And then everything takes so long that the answers can’t come in fast enough to meet business needs.

One example of what a CRE company can achieve, he said, is a client in the affordable multifamily space. The cost of land is a brutal constant. Someone at the business had a brilliant idea to look for parcels owned by a hospital, university, or similar organization and then to see if they could partner with them to develop housing with revenue-sharing rather than purchasing. “It’s really hard to do this,” Salmanson says. “It would take analysts months. It’s a two-second query in Cherre.” But it’s not something Cherre would come up with.