Ok, so he’s not totally wrong. We DO need a tradable data marketplace, but I totally disagree with the premise of how we get there, and who’s going to do it. As CEO of the largest data marketplace in our industry – operating in 60 markets and with 15,000+ members, you can imagine that I have some strong opinions on the subject.
As a refresher, here’s how Duke laid out his utopian data marketplace. I’ll address each bullet point individually.
A FREE secured database populated by (with the contributing companies as partners) public and readily available data.
I agree FREE is great! It seems only fair that if you’re a data contributor, you should be able to get data out for free. In fact, we modeled CompStak Exchange off of this premise – CRE brokers, appraisers and researchers share comps on CompStak, they earn credits for sharing those comps and can use those credits to get other comps back out. It’s totally free for our Exchange members.
However, as your parents probably told you when you were a kid, “nothing in life is free.”
So who’s going to pay the expenses for Duke’s utopian exchange? Will Cresa or Avison Young be willing to pay as much as CBRE? More importantly, will CBRE allow Cresa or Avison Young to have access to all of the same data? I doubt it.
CompStak allows brokers, appraisers, and researchers trade for free on CompStak Exchange, but the exchange still needs to be built, maintained, supported, debugged, and data needs to be processed and hosted online. All super expensive. This is why we charge the world’s largest owners, investors, and lenders (like the owners Duke mentioned in his post) a fee for access. CompStak is always free for brokers, appraisers, and researchers.
Since the inception of the internet, many industry consortiums for trading data have come and gone. Outside of very small local players, these consortiums have largely failed. They’ve failed, because organizations that are NOT technology companies are trying to pretend that they are, all while navigating the politics of making all of the major CRE firms work together. These projects are mostly doomed from the start.
The database will have an agreed to standard base data set.
Again, I agree with this in theory, but again, it’s a utopian fantasy. An agreed standard base data set would be fantastic, but expecting the major players in CRE to all agree on a standard and then apply that standard to their data is wholly and totally unrealistic.The data tracked by CRE firms is in PDF documents, Excel, Word, Emails, ancient custom built CRMs and the notes on a broker’s desk. To make matters worse, the way data is tracked varies market by market and even document by document!
A few examples:
-Many owners track brokerage commissions in their Net Effective Rent calculations, but brokers don’t.
-Rents in San Francisco are tracked per SF per year, but just outside of San Francisco in the greater Bay Area (and most of the rest of California for that reason), rents are tracked per SF per month!
-In San Francisco, tenant improvement allowance is not included in the Net Effective Rent calculation, but in New York City it is.
-In much of Europe, rents are quoted per meter. In the UK, they’re still per square foot.
So how do you overcome this data standardization issue? In short . . . you don’t. At CompStak, we take data in any format, following any convention (or no convention at all). We process over 50,000 comps per month with dozens of data points each, and we receive this data in PDFs, Word documents, and in the body of emails. We’ve even had people mail us comps! We then standardize the data on our end. However, we present the data to our users in whatever format is most common for the market. If you’re in San Francisco you’ll see your rents per SF per year, and if you’re in Palo Alto, you’ll see them per SF per month.
The data will be updated by the entire marketplace.
Let me introduce you to the 90-9-1 principle – https://en.wikipedia.org/wiki/1%25_rule_(Internet_culture)
It says that in an internet community (or crowdsourced platform) 90% of participants view only, 9% edit content, and only 1% of participants create content. At CompStak, our contribution rate is far higher, but the rule still stands, most participants in a crowdsourced community don’t contribute.
Each contributor will gain a Value for that contribution.
Agreed. In this case what’s even more important is that the value you take out is proportional to the value you create! This makes sense in theory but is hard to make work in practice. At CompStak we’ve spent years and a lot of tech dollars to optimize a system that works.
The value is created with the database and stays within the database.
Ok. I think I get this, but you’re leaving a lot of value on the table if you’re just thinking about the data as a database, and not using it in other ways.
They will NOT pay to source other contributors data.
See above regarding (immense) costs of operating an exchange (effectively).
The source of and verification of that data will be the benchmark for inclusion in the marketplace.
In this utopian free marketplace, where everyone contributes and no-one pays, who’s checking the sources of the data and verifying it? How do you keep everyone honest? How do you create “benchmarks?”
So in summary, why is Duke wrong? He’s wrong because of his plan, like socialism, is better in theory than in practice. He suggests that CRE professionals are better off creating technology than a “software vendor” or “tech executive.” That is EXACTLY the kind of foolish thinking that has kept our industry in the stone age. CRE brokers should focus on brokering deals. CRE investors should focus on buying buildings. Or if you’re really passionate about this BECOME a CRE “tech executive” . . . that’s what I did.
By the way, I’m NOT suggesting that commercial real estate owners, brokers, lenders or others in the industry should let CRE Tech companies control their lives. At the end of the day, the professionals and companies providing the data should and DO have immense control over the way this data is shared. However, as a CRE professional you should spend your resources supporting the technology companies that 1) make the exchange of information fair, and 2) treat their data contributors and customers fairly.
Ultimately, if there’s a company that profits by creating a fair marketplace, then there’s financial incentive to ensure it works. In the battle between free private enterprise and consortiums, free enterprise always wins.
Why Duke Long Is Wrong. There Is a Tradeable Data Marketplace. A Rebuttal: Michael Mandel.