You: “Where should I price this house?” Me: “How much are the buyers willing to pay?” You: “How am I supposed to know about the buyers?” Me: “Precisely!”

For a year-and-a-day, I was an iBuyer pricing algorithm.

When Zillow announced they were going into the cash-incineration iBuying business, I looked for opportunities to find out more. Amazingly, I lucked into a job doing CMAs for Zillow’s pioneer broker on the ground. What’s amazing? Until then, I had never done a CMA in my two-decade career as a Realtor.

I could and can price to the dollar, but I never did anything beyond reading the comps like tea leaves. Zillow was a huge gift to me that way. I got to work with appraisers, and I learned why their comp selections always look so alien to me. I did my best to teach them my way of comping, but I only made a real dent in two of them.

Their tricks were great, and I had fun playing with them. I had never in my life adjusted for anything, for one thing, but I had never ever crossed an arterial street to get a comp before, either: It was – and still is – a great big deal to get me to leave the subject property’s subdivision.

My biggest influence on them was time. Until March of this year, the price trajectory in the Valley of the Ever-Fecund Sun had been steady as a clock since 2014 or so. They were going back three months for comps, adjusting for time. I rocked them to a year, and I did not hesitate to go back longer than that. Better to have closer comps, adjusted, than a false confidence in more-recent but less-comparable sales.

Zillow was a hoot. The appraisers would do things I think are crazy – like comping one- to two-story homes – but Zillow’s minions would not understand why wildly divergent product lines – Ikea kitchens versus Sub-Zero kitchens, for example – cannot be compared for pricing purposes. If you’re looking for defects in any AVM, it’s there, the inability to reconcile proximate but dissimilar parcels. (I know how to fix that, for what that’s worth. Hint: The best comp to any residential property is itself at prior closings.)

Meanwhile, my differences with their praxis, which I taught to them with limited success and to Zillow’s people with somewhat better impact, can be expressed in one sentence:

You’re not comping the house, you’re comping the buyers.

The goal is to determine what the buyers of this house will do, and the answer to that question is to be found not by looking for similarly-priced homes but, rather, for similarly-motivated buyers.

That’s why I hate to leave the subdivision or master-planned community, and it’s why I’m willing to adjust for time to get model-matches – or the next-best-thing – to try to suss out my buyers. It’s why I never had to do a CMA, too, and why I never had to adjust for anything: I don’t care how much you want to fudge-factor for a comp with a pool if the subject property doesn’t have one. Why? Because the other buyers didn’t want a pool, either, and I want to know about those folks. They’re the ones who know what the next buyers will do.

I’m talking about bread-and-butter houses, but the same principles apply to luxury, historic, architecturally-significant or other “incomparable” – meaning uncompable – properties. As either the listing or buyer’s agent, your objective is to determine what the buyers will do – and then hope that the appraiser can see things the same way. The difference would be, iBuyers have no business screwing around with jewelry-box homes, and yet there they are…

I came out ahead, anyway. I got to a place where I could price houses unassailably. Getting to the number was always easy, but I was in a zone where I was Delphic: Here are four model-matches in the sweet spot, plus two outliers to show you why it’s the sweet spot. It was life in a batting cage, and there is no better way to get good at putting the ball out of the park again and again. I was living my regular life and doing my own work throughout, but hard work is where I dance. In January of 2019, I slept less than three hours a day.

But in the end I came back to where I started: Most-comparable, most-proximate, most-recent – in that order of priority. Close comps, the closer the better, so you can clearly see the buyers’ needs that will sell that particular home. Adjust for as little as possible – but don’t lose the buyers. They’re the only people who know for sure how much they’ll pay…