Using sold listings on Zillow for comps

I’ve been using Zillow to analyze land prices. I assume prices for properties listed as sold to be more accurate representations of full retail price and for generating comps. So I tend to look there rather than at the for sale prices. But now I wonder if there is an important difference between the for sale listings and the sold listings on Zillow that’s not obvious.

I assume properties listed for sale on Zillow are by definition listed properties, as in listed on the MLS database. But when they list sold properties are they adding properties reported as sold by the county that were never listed on the MLS system? This would skew prices in the sold list (on average) more toward unlisted property sales and away from full retail price.

Here’s an interesting property I found on Zillow (see link below) which sold two years ago for a surprisingly low price that seems to suggest what I refer to above. This is a 3 bedroom home on nearly two acres in Northern California that sold for $10,000. That’s about 3-6% of market value. This does not look to me like a sale of a property that was ever listed on the MLS.

https://www.zillow.com/homedetails/15280-Whispering-Pines-Dr-Shasta-CA-96087/15203372_zpid/

Is there a fundamental difference between the prices shown on Zillow’s for sale listings and the prices in their sold listings?

@landoffer In my experience I would suggest sticking to comps in and around the the most common parcels. It’s it’s way outside the average set of comps be it too cheap or too expensive then I discard it. Also a good idea to read the description to see if there’s any improvements such as fencing or small cabin, water or electric hookups etc. If you are selling completely vacant land then you can’t really compare to anything with improvements. OR be sure to discount your property for lack of those improvements. Just focus on the getting the average of the most average and you should be close.

@push4ward Thanks for the sound advice! So much to learn. Rereading my original post, I might have asked the question differently.

On Zillow there are two sets of data that can be used for comps, the For Sale and the Sold listings. Initially, I was looking at the prices in For Sale as asking prices, which is some indication of what retail prices are but I would think somewhat inflated. Amounts listed in the Sold section I considered a better reflection of what actual retail prices are in the area I’m looking at. In the sale of goods this might be referred to as the “street price”.

But then I noticed there are usually a number of properties in the Sold section with dramatically lower prices. It occurred to me there must be something beyond ‘for sale’ versus ‘sold’ about the two lists.

My original question in this post is whether the Zillow Sold listings come from county records and the For Sale listings are generated by agents and private sellers. Since asking the question here I’ve done more research and confirmed from Zillow itself the Sold listing include sale prices published in county records (https://zillow.zendesk.com/hc/en-us/articles/214908558-How-does-Zillow-receive-price-history-information-). I’m now pretty sure that the much lower amounts you see paid for some properties in the Sold listings are sales to land investors like ourselves and should be thrown out when researching comps.

This may have been obvious to everyone but me, but I’m also asking to find out how others are using Zillow for comps. Do people ignore the prices in Zillow For Sale listings completely? Using your process of throwing out the high and low outlier amounts, do you only apply that approach to the Sold listings only? Or do you have a way of including asking prices from For Sale listings in your analysis?

Perhaps someone reading this can share their step by step process for deriving vacant land comps from Zillow listings.

@landoffer Was the example you cited an arms-length sale? It could have been a sale between close relatives? It might have been a sale to settle a debt owed to the buyer.

@landolease I thought of those possibilities, yes. It’s another reason to ignore the outliers when looking at comps. I cited the example because it was seeing that sale that alerted me to there being non-standard transactions included in the Zillow Sold listings. I’m obviously still figuring this stuff out. Thanks.