There is a house on West Wackerly Street in Midland, Michigan. It went up for sale, someone bought it, and the sale closed. Nothing unusual.
On almost any other website, the page for that house would be gone by now. The thing it was selling no longer exists, so the page gets deleted or redirected, and whatever it had earned in Google disappears with it.
Zillow still has the page.
Not as an archive, not as a redirect — as a normal, live page that simply stopped being for sale. And when we counted, we found that homes like it outnumber the homes Zillow actually has on the market by a wide margin. The pages that should have died are the biggest thing on the site.
This is the story of why Zillow keeps them, and how you can do the same thing with your own expired pages next week.
Results at a glance
A home on Zillow moves from for sale to pending to sold to off market, and it keeps the same web address the whole way through.
Nothing is ever deleted. The offer ends; the page stays.
The house that never left
Think about what a listing page actually is. Most of it has nothing to do with the sale.
The address is permanent. So is the neighbourhood, the school district, the year it was built, the number of bedrooms, the tax history, and the price it last changed hands for. The only genuinely temporary parts are the asking price and the "contact this agent" button.
So when the house sells, Zillow does not have a dead page. It has a page that lost two elements and kept twenty.
And people still look for it. Someone wants to know what the house down the street went for. Someone is researching a block before making an offer. Someone types a full street address into Google because they are curious. All of them land on a page that has been sitting there quietly for years.
Most sites delete. Zillow relabels.
Here is the habit worth breaking.
When something expires, most teams treat it as a row that has been removed from a database. The job was filled, so the job page goes. The product was discontinued, so the product page goes. The event happened, so the event page goes.
Zillow treats expiry as a change of status, not a deletion. The house is still a house. It just is not for sale any more.
That sounds like a small distinction. It is actually an architectural one, and you can see it in the web address:
/homedetails/2417-W-Wackerly-St-Midland-MI-48640/
The address. Not the listing, not the price, not the agent, not the year it was sold. Nothing in that URL has to change when the sale closes — which is exactly why the page can survive it.
If your own URLs contain the thing that expires — a season, a price, a job number, a batch — you have built deletion into your own website without meaning to.
5,000 sitemaps versus 35
You do not have to take Zillow's word for any of this. Sitemaps are public, and they give the whole game away.
Zillow keeps a separate set of sitemaps for each stage of a home's life. One for homes for sale, one for homes under offer, one for homes recently sold, one for homes off the market entirely. Sorting them by size tells the story in a single picture.
The inventory that makes Zillow money is the sliver. Everything above it is history.
And that history keeps growing. Homes come onto the market for a few months and leave it forever, so the archive only ever gets bigger while the for-sale section stays roughly the same size. Zillow built a website where the passage of time is an asset instead of a maintenance problem.
Fewer links than you would expect
Now the part that surprised us, and the part that makes this strategy usable by someone who is not Zillow.
We ran Zillow through our own Backlink Analyzer expecting an enormous link profile to match the enormous page count. It came back smaller than Booking.com's, and smaller than Canva's.
Which makes sense once you say it out loud. Nobody links to the page for a house that sold seven years ago. Those pages were never built to attract links, and they never will.
They were built to be found — by one person, typing one specific address, looking for one specific answer. Multiply that by every home in America and you have something no amount of link building could buy.
This is the encouraging part. An archive like this does not require a marketing budget or an outreach campaign. It requires you to stop throwing pages away.
Schools outlive listings
There is one more piece, and it is the piece most people skip.
Alongside all those homes, Zillow keeps pages for things that can never expire at all: schools, school districts, neighbourhoods, local landmarks, mortgage rates, home values for a city.
A school does not sell. A neighbourhood is never taken off the market. So this layer sits underneath the listings and holds the whole area together, no matter how many individual houses come and go on top of it.
If you only preserve expired pages and give them nothing to sit on, you get an orphanage. Pair them with something permanent and you get a section of your site that stays alive between refreshes.
Copy this in an afternoon
You do not need Zillow's data or Zillow's brand to use any of this. Four steps, in order.
1. Find the pages you are currently deleting. Sold products, filled jobs, past events, last season's collection, closed listings. Write the list down. That list is the opportunity.
2. Ask what is left when the offer is gone. For each type, cross out the temporary parts — the price, the button, the availability — and look at what remains. Specifications, photos, a location, a date, a result. If something real remains, the page deserves to stay.
3. Change the status instead of deleting. Keep the page at the same web address and change what it says. "Buy now" becomes "this sold in March for $340,000." A filled job becomes a record of what that role paid. Same URL, new state.
4. Give the archive something permanent to lean on. A category page, a neighbourhood, a brand, a topic hub — anything that will still make sense in five years. Link the retired pages to it.
That is the whole strategy. It costs nothing, it works on a site with fifty pages, and the main thing standing in your way is the instinct to tidy up.
What you cannot copy
Zillow's raw material. Property records in the United States are public, so a sold home still has real information attached to it. If deleting the offer leaves you with an empty template, preserving it is just clutter with a longer lifespan. Redirect those instead — be honest with yourself about which is which.
Zillow's reputation. A giant, barely-linked archive works because the rest of the site is trusted. A brand new website preserving everything it ever published is making a very different bet, and should grow the archive slowly.
And here is what we could not check ourselves, said plainly:
We never saw a Zillow page. Every time we asked for one, Zillow refused us — from a simple script and from a real browser alike.
A page in a sitemap is a request, not a promise. Zillow is telling Google these pages exist. We have no way of knowing how many Google actually keeps, and some of that archive is certainly crawled rarely and lightly indexed.
We have no traffic or revenue figures for Zillow, and this article does not claim any.
Before you delete anything, find out whether it is carrying links. The same Backlink Analyzer report shown above works on any website you like, and it takes about a minute.
- The same check we ran on Zillow
- Referring domains, anchors and the follow split
- 3 checks a day, no signup, no card
Frequently asked questions
How many sitemaps does Zillow have?
Its robots.txt declared 15 Sitemap: lines when we read it on 21 August 2026. Nine of them are home pages split by listing status, one covers things that never expire such as schools and home values, and the rest cover buildings, apartments, city pages, search results and a separate mortgage section. There is no sitemap at the usual zillow.com/sitemap.xml address — that returns a 404, and everything is declared through robots.txt instead.
How big is the archive, exactly?
The off-market section lists 5,000 sitemap files. We opened six of them, spread across the whole range, and each held between 23,427 and 23,747 web addresses. That works out at roughly 118 million pages. It is an estimate from a sample, not a figure Zillow publishes, and declared pages are not the same as pages Google has indexed.
Why does Zillow have more sold homes than homes for sale?
Because homes arrive on the market temporarily and leave it permanently. Only a few million American homes are listed at any one time, while every home that has ever sold stays in the database. Zillow keeps a page for each, so the archive grows steadily while the for-sale section stays about the same size.
What is Zillow's backlink profile?
Our Backlink Analyzer returned Domain Rank 94 out of 100 and 344,466 referring domains on 21 August 2026, with 85% of links marked dofollow. We ran it twice and every figure reproduced except the headline backlink total, which moved between 34.2 and 33.4 million across the two runs, so treat that one as approximate.
Is that a big link profile for a site this size?
No, and that is the interesting part. It is the smallest of the three large brands we have measured with the same tool — Booking.com came back at 1.38 million referring domains and Canva at 624,000, against Zillow's 344,000. Spread across a site this large, that is roughly one referring domain for every few hundred pages. The archive was built to be found in search, not to attract links.
Should I really keep my expired pages?
Only when something useful survives the expiry. Zillow can keep a sold home because the address, the school district, the specifications and the sale price are all still true. If removing the offer leaves you with an empty shell, you are preserving thin content, and redirecting it to the relevant category is the better move.
How did Zillow score in a technical audit?
88 out of 100 on our Website Audit, with no critical issues and 112 checks passed on the homepage. The three warnings were dead outbound links, an oversized image and images missing alt text — the first of which is a little ironic on a site built never to delete a page.
Can I check any of this myself?
The sitemaps, yes. Open any website's robots.txt, find the Sitemap: lines, and follow them — it is all public and it takes about ten minutes. Zillow's own pages, no: it refused every direct request we made, including from a real browser. Our tool figures come from databases that never need to visit the site.
Decide what deserves to stay
Every company in this series answered the same question differently. RTINGS adds measurements, Wise adds currencies, Zapier adds pairs of apps, Canva adds templates its users made, Booking adds new ways of wanting a hotel.
Zillow is the first one that had to answer a harder question: what do you do when the thing you built the page for goes away?
Its answer is that nothing goes away. The sale closes, the button disappears, and the house on Wackerly Street keeps its page.
So the question for your own website is not how many pages you can create. It is which of your pages would still be worth reading the day their reason for existing expires — and whether you are about to delete them.
Before you do, check what they have earned. It takes a minute, and you cannot get a page's history back once it is gone.
Measured on 21 August 2026. Sitemap figures come from zillow.com's robots.txt and its sitemap files, fetched directly. The 118 million estimate is based on a sample of six files out of 5,000, with the counts published in the FAQ above.
Link, keyword and audit figures are free runs of our own tools; exact values are in the FAQ. Pages listed in a sitemap are not the same as pages indexed by Google, and no traffic or revenue data for Zillow was available to us or is claimed here.