I don't understand why this is necessary, or why asking someone who hasn't been on the housing market for 15 years what the value of their home is might be good data collection.
I bet if you asked a panel of 1000 homeowners that haven't bought or rented in 10+ years what rent was going for in their cities, they'd undershoot it by 20% or more.
> the CPI views housing units as capital (or investment) goods and not as consumption items. Spending to purchase and improve houses and other housing units is investment and not consumption. Shelter, the service the housing units provide, is the relevant consumption item for the CPI. The cost of shelter for renter-occupied housing is rent. For an owner-occupied unit, the cost of shelter is the implicit rent that owner occupants would have to pay if they were renting their homes.
If I'm reading that fact sheet correctly, they actually don't ask homeowners at all. Rather they just ask a bunch of renters whet their monthly rent is (which the vast majority of them should be able to answer accurately) and then use fancy math to extrapolate what the nearby home ownership units would rent for.
I'd suggest reading/skimming that whole fact sheet if you want more details, though the summary seems to be that they've actually put a bunch of thought into this stuff.
[0]: https://www.bls.gov/cpi/factsheets/owners-equivalent-rent-an...
But imputed rent is a more subtle and difficult number. The goal is to know what the market rent would be for all the homes not currently on the rental market. I'm sure you can find other sources on why this value should be included in the CPI calculation (in addition to rent), but once you recognize that, it should be clearer why the Zillow approach wouldn't work: Zillow and similar sites only know about what homes sell for, or rent for.
Now could an expert come up with a model of what a home might rent for given sale price, number of sqft, bedrooms, etc.? Of course. But doing so very much isn't an afternoon project because the data is tricky: the owner occupied homes differ from rental units in important and hard to evaluate measure. For instance, apartments around my city are more poorly maintained than similarly aged ownership units.