[VIP] Zillow and Opendoor Q3 iBuyer Data: Denver

In the last Phoenix iBuyer comparison, I said I was compiling some data out of Denver. Well, here it is for your enlightenment and entertainment.

One thing to note: Offerpad is not active in Denver. Redfin is, but my dataset has no Redfin sales. I hope to see if I can get my hands on some RedfinNow data, as Redfin remains one of the most interesting iBuyer companies with zero data available.

Let’s see what Denver can tell us about Q3 iBuyer activities.

The Comparison: Denver

Here’s the Google Sheets spreadsheet. I’ve been told that you can’t just click through, so if you need a link to the sheet for your own analysis, contact me. I’ll do that for my VIP subscribers.

Let me reiterate some of the caveats from other posts.

What we know is that Zillow has said that it asks for Seller Concessions that are equivalent to their estimated Renovation Costs. Zillow at least does not use Seller Concessions as a profit center. We don’t know if Opendoor does the same or if they use Seller Concessions as a way to squeeze out a few bucks of profit.

Holding Costs — we have Zillow’s national numbers for Q2, but I took the step of trying to estimate holding costs for Denver metro using averages for things like utilities, HOA fees, property taxes. Obviously, that is not specific property-level holding cost data, but… it’s something.

Renovation Costs — in a way, I don’t think this is important since at least one of the three companies clearly states that its Seller Concessions = Renovation Costs. So they net out in a way. But I will note that we only have Zillow’s national numbers from quarterly reports; we’ll see what they report for Q3.

So, what takeaways, if any, are there in this dataset from Denver?


First, the Phoenix Q3 data showed a domination by Opendoor. I wrote in that post:

While Zillow and Offerpad both sold fewer homes in Q3, and quite substantially fewer, Opendoor sold 17.4% more. In Q2, Zillow sold 330 to Opendoor’s 574, or 57.5%; in Q3, those numbers are 299 to 674, or 44.4%.

Similarly, I estimate that Zillow’s revenues went down by almost 8% while Opendoor’s revenues skyrocketed by 20%. At least part of the reason for the big difference in revenues between these two has to do with Opendoor’s average profit per trade going up by 6.2% to $9,312, while Zillow’s fell almost 90% to $249. Is that intentional by Zillow to try to get more properties? Or is it that Zillow’s algorithms were not doing all that well in Q3? I don’t know, and no one outside of Zillow does. But it is striking.

Whatever the reason, I think it’s clear that Opendoor dominated Phoenix iBuyer market in Q3.

The Denver data does not show that kind of domination. If anything, we see some major slips by Opendoor, which makes me think that they might have made some real mistakes in pricing these particular homes.

Zillow sold 100 homes in Q3 vs. 113 by Opendoor, but in Q2 those numbers were 55 and 102 respectively, which says that Zillow has almost caught up to Opendoor. What’s more, Zilow’s total Volume is up 77% from Q2, while Opendoor’s Volume slipped 50bps.

Both companies saw the average sold price slip, which might hint at some softness in the Denver housing market, but Zillow’s slip was only 2.7% vs. Opendoor which saw average sale price drop by 10.2%. And for both Zillow and Opendoor, the average DOM was up — again, implying softness in the Denver housing market.

The data that jumps off the page, however, is Opendoor’s poor performance in Q3. The average Sale to Purchase dropped below 100% for the first time in any market I’ve looked at, ever. That means Opendoor actually lost money on each trade, before taking fees into account. From an average profit of $4,900 in Q2, Opendoor lost $6,550 on each trade in Q3. That’s a big drop. Accordingly, topline revenue for Opendoor was down 36% and gross margins (which takes seller fees into account) dropped by 40%. Ouch!

Looking at Opendoor’s data in a bit more detail, I’m seeing that it only “won” on only 43.4% of its trades. In comparison, Zillow won on 90% of its trades, even if the average margin slipped from Q2.

Consumer Impact

Maybe because Opendoor had such a bad quarter, the Consumer Impact analysis goes the other way for Denver.

In Phoenix market, it was striking just how much better Zillow was for sellers:

The average seller would have net a measly $2,496 for an additional 3.7 months on the market had he chosen to list with a REALTOR over selling directly to Zillow. But that amount is $10,263 for Opendoor and $12,943 for Offerpad — and that’s without knowing the realities of what those companies demand in Seller Concessions, for 3.1 and 2.8 additional months on market.

In Denver, it’s the exact opposite. The average seller would have net $9,880 more had he chosen to list with a REALTOR over selling to Zillow, but would have lost an additional $4,978 had he done that instead of selling to Opendoor. So, for most of the Denver sellers who sold to Opendoor in Q3, congratulations! You likely saved yourself three and a half months of inconvenience and made an additional $5K.

Why the Difference?

Here’s what I’m pondering now. Why the difference between Phoenix and Denver between these two companies?

There’s no doubt that Zillow’s Q3 performance in Phoenix was… if not disappointing, then at least against the trends of the past few quarters. In Denver, where Zillow has only been active since Q1 of 2019, we see near doubling of transactions, very healthy growth, and decent profitability (even with what looks like a market slowdown).

Similarly, Opendoor’s Q3 performance in Phoenix was impressive, but it really screwed the pooch in Denver.

The question is, Why?

Nobody outside of Opendoor and Zillow knows, of course, but I think there are three possibilities.

One, Opendoor just screwed up pricing in Denver in Q3. It’s a one-time thing, could happen to anybody, and Opendoor adjusted its algorithms or buy box or whatever. Looking closer at the data, you can almost spot when the company has made a mistake or the market has shifted on a company, followed by an adjustment.

Two, Zillow pulled back in Phoenix in Q3 for some reason. Maybe Zillow didn’t have the right kind of infrastructure in place, had a hiccup somewhere with execution, and decided not to lose a bunch of money and pulled back on purpose. That makes some sense to me because Zillow’s margins, revenue, etc. weren’t that bad; it’s profit-per-trade was way lower, but… it made a small profit, while selling fewer homes.

Is it the case that Zillow tries to grow as quickly as possible up to a point, then hits a wall? Should we expect similar patterns in all of its iBuyer markets?

If so, we should see a similar pullback in Denver in a couple of quarters. Time will tell.

Three, Opendoor is responding to competitive pressure from Zillow in Denver and overpricing homes a bit in order to win them, which led to the Q3 performance. If so, there are reasons to be concerned for Opendoor.

The competitive advantage that Zillow has is structural: its lower CAC due to the consumer portal. It will always be the case that Zillow can bid more for a property, sell for less, and buy higher priced homes, because its CAC is so much lower than anybody else’s.

In effect, Opendoor (and Offerpad and anybody else) either has to be so much more efficient than Zillow to overcome the CAC disadvantage, or neutralize the CAC disadvantage by either (a) word of mouth, or (b) some other marketing network, such as real estate agents. None of those appear promising right now but again, I suppose time will tell.

That’s it for now. I hope to compile all this for a report on the state of iBuyers sometime in 2020 after Zillow and Redfin have reported for FY 2019. If you have any thoughts or questions, sound off in the comments, or in the Notorious VIP Lounge.



A note about the video: Sunny and I were at this concert, and it was one of the most impressive live shows we’ve ever attended. If ZBB comes by your area, I highly encourage going. 🙂

Share & Print

Rob Hahn

Rob Hahn

Managing Partner of 7DS Associates, and the grand poobah of this here blog. Once called "a revolutionary in a really nice suit", people often wonder what I do for a living because I have the temerity to not talk about my clients and my work for clients. Suffice to say that I do strategy work for some of the largest organizations and companies in real estate, as well as some of the smallest startups and agent teams, but usually only on projects that interest me with big implications for reforming this wonderful, crazy, lovable yet frustrating real estate industry of ours.

Get NotoriousROB in your Inbox

13 thoughts on “[VIP] Zillow and Opendoor Q3 iBuyer Data: Denver”

  1. ROB,

    I feel I’m pretty in tune with our business. But, I never understood the appeal of the iBuyer model and still don’t (from the business side – I understand the consumer benefit). Is it to create large increases in revenue, the mortgage side, selling leads? It can’t be flipping homes for a profit at scale, and making consumers happy isn’t a business model.

    So, just how large is this market today? My gut tells me it’s a lot smaller than perceived….

    We’ll see.


      • I read it. If I recall they used the total addressable market – something over a trillion. I’m interested in knowing how many homes are within their buyer parameters today. Houses built after 1980, priced between $200,000 and $450,000 that are located in neighborhoods where the housing stock is commodity-like.

        Personally, I’ve only seen those communities in picture so I have no personal experience in seeing where all these iBuyer opportunities exist and how many there are. I’m convinced it’s a very narrow pool or at least more narrow than how the market is treating the model.

        It should be a simple number to come up with….I’m thinking the iBuyers have an actual count.

        Hey, I may very well be wrong – but just looking for a number.

        Size matters (in this case anyway) 🙂


      • I realize there is a difference. But let’s start with today, I would think it is important. I’m not a believer that the model has viability in housing markets with unique housing at the top end of their price range and above.

        Once they get into those higher numbers, a Home Depot make over doesn’t fly. As the numbers move even higher renovation becomes even more difficult and serious in terms of risk. More often those renovations compete against new infill construction and really, really smart local players as competitors.

        We’ll see.


      • So Brian, it seems to me that the real question to be asked is, “What percentage of American housing stock is in markets with unique housing at the top end of their price range and above?”

        I’m gonna go with “less than 40%” and stick with my 60% market share prediction.

  2. How about option #4? Redfin & OD have a partnership in Phoenix & Redfin is sharing their brokerage data w/ OD creating a major strategical advantage.

    • That is a possibility, I suppose. But pretty sure the Redfin-OD deal is nationwide, not just Phoenix. And Redfin is in Denver, which makes that a bit suspect.

      FWIW, Redfin is the only other company that has the lower CAC that Zillow has. And if the partnership with OD did not involve OD paying Redfin a referral fee, I’d suppose we could say that neutralizes Zillow’s structural advantage. But it does involve paying a fee, so… advantage Zillow (and Redfin, except Redfin doesn’t publish numbers).

  3. ROB,

    Sorry, couldn’t let it go….

    Looks like we need to consider the housing stock below $250,000 in that equation 😉


      • Drew,

        Exactly! The circle is getting smaller.

        For transparency; I have nothing against iBuyers or their model. I hope they do great. My interest is in how and why a trend like this can catch fire so fast with so much participation without a common understanding of what’s going on.

        We’ll see. 🙂


Comments are closed.

The Future of Brokerage Paper

Fill out the form below to download the document