How accurate is my Zestimate?
One of the first questions we’re always asked when talking with a potential client or even a friend when real estate comes up, is: “Do you think my Zestimate is correct”?
before
after
In a word, no.
How could it be? It’s only an algorithm based on the available data. Furthermore, your home is only worth what someone is willing or capable of paying for it. How does an algorithm calculate this very human factor? {More on that below.}
Zillow has good data, but what they do with that data is not always the most beneficial. We’re not picking on them. It’s true for any online RealEstate site.
They do their best, but their best is not the best for you.
before
after
For example, I can program a robot to cook some ribs, but I’d be willing to bet you’d rather have a seasoned grill master who is hunkered over that grill every day of the week, instead. An actual human expert can take into account varying factors of the slow cooking process, and make adjustments for factors that a robot couldn’t understand, because it isn’t in the algorithm. Lame example, but maybe it helps.
Let’s try this example instead… and break down a standard Zillow Zestimate example.
As we do, notice the factors of each house. For reference, these two houses are in the same neighborhood and 0.35 miles apart. From a comparables standpoint, they’re ideal.
We’ll call the house on the left “Subject A” and the house on the right “Subject B”
The first question that should jump out at you is, why is Subject A so much less than subject B?
After all, subject A is a larger home with more bedrooms. In fact, Subject A has almost 1500 sqft more with an extra bathroom.
You first guess may be that perhaps it is that subject B was renovated and Subject A is an older home. That, after all, can be a major factor in home values. When you pull renovation records, Subject B’s last major renovation was completed Dec. of 2022, so that’s pretty recent. Perhaps the answer is simple, and the Zestimate is right.
Until you pull Subject A, and find out they completed major renovations in June of 2022. So do 6 months make that kind of difference? If you believe it does, then as the saying goes “I have some ocean front property in Kansas for sale”.
So what gives, Zillow?
In simple terms, the algorithm can’t make real-life adjustments like a good Real Estate agent can.
Sitting directly across the street from Subject A is a house (Subject C) that was purchased by a person who didn’t have a clue what they were doing. They started a renovation without pulling a permit and were busted by the codes department. Long story short, they’ll be waiting for 18+ months to get a permit from the day they purchased the home. Now, they’re trying to unload the property and get out from under it because they know they made a bad decision. They got in over their head, and they messed up. Zillow doesn’t account for that, and so many other very human things.
So it {in Zillow’s current comps} artificially devalues the homes around Subject C.
This is because House C is back on the market, completely gutted out amongst many other things, for a very low per sqft rate. {See unloading debacle, above.} But this sqft rate is only because the person who is selling Subject C got “caught with their pants down” for lack of a better phrase. Not because it reflects the actual finished value of anyone’s property.
Does that mean the properties around Subject C are any less valuable than Subject B? Again, if you believe that, hello beachfront in Kansas.
If I listed Subject A today, I would list it at 1.28 Mil and expect it to sell for 1.35 mil+. It’s in a thriving neighborhood with a great location, one that is highly sought after. It has lots of great developments around it. The list honestly goes on and on. Subject C will, in time, be corrected by someone who knows what they’re doing.
This brings me back to “willing and capable”.
Another very human factor that the algorithm doesn’t understand. What I know is that the last few homes that sold in Subject A’s neighborhood, both sold over asking and were under contract within 48 hours. Any listings have crowds of cars lining the streets. Competition is still stiff in this area. So as an agent {not an algorithm} I set a slightly lower-than-expected list price, understanding that it’s likely I will receive multiple offers, thus why my expectation is a selling price of 1.35 mil +.
So back to my original statement: Zillow has good data, it’s just not always beneficial.
It’s kinda like trying to google your symptoms on the web. You might find good data, but is it beneficial? You really need a professional {aka a doctor} to be properly diagnosed. And a chef for those ribs, too.
Real Estate transactions are probably the single biggest transactions you will ever make, so don’t be foolish {like the person who never should have purchased house C}. Get advice from a trusted and educated REALTOR in your local area. If you don’t know one, ask us. We probably do! We don’t want to see you make a mistake whether you’re buying or selling your home.
Thanks for tuning in with us today while we demystify Zillow. Be sure to check out more of our real estate posts here!
If you’re in the Middle Tennessee area, contact us here to work with us + here to talk all things renovations! If you’re not familiar with what we do, check out this post here for more info.
Ps. These photos are from the blue house project – be sure to check out more + some of the inside, here!
Have an inspired day!
Leave a Reply