Check this page with an assistantOpens a chat asking it to summarise this article and name the evidence behind each claim.
Claude opens with the prompt on your clipboard: Anthropic does not support prefilled prompts on the web, and we would rather copy it than ship a button that drops it.
The Feed Problem, Stated Plainly
Listing data comes from a shared source and appears, near identically, on every agent site in the market. Your listing page for a property is one of hundreds of copies with the same description, the same photos, and the same specifications. There is nothing on your version to prefer.
Two further properties make it worse. The pages are thin by construction, because the feed supplies specifications rather than substance. And they expire: a property sells, and the page that was accumulating anything stops being relevant within weeks.
Set against portals that carry the entire market and have been accumulating standing for years, this is not a contest to win by optimisation. It is a structural disadvantage, and recognising it early redirects effort toward ground you can actually hold. That is a selection judgement rather than a technical one, of the kind described in search intent.
Deciding What Should Be Indexable
A site generating thousands of near-duplicate pages that expire is spending crawling on pages that will not rank and diluting what the site appears to be about. The question is not whether visitors can search listings, which they must, but which of those URLs should be indexable.
| Page type | Indexable? | Reasoning |
|---|---|---|
| Individual feed listings | Usually not, at volume | Thin, duplicated, short-lived; they will not win |
| Your own exclusive instructions | Yes | Unique to you, worth original description and photography |
| Search results with filters applied | No | Combinatorial; see faceted navigation |
| Neighbourhood pages | Yes, these are the asset | Original knowledge, permanent, no feed equivalent |
| Market reports | Yes, on one updated URL | Compounds annually rather than fragmenting |
| Agent profiles | Yes | People search agents by name; see ranking for your name |
The exclusive-instruction row is the exception worth protecting. A property only you are marketing is genuinely yours to rank for, and it deserves original description, your own photography, and a proper page rather than feed output. Those are the listing pages worth treating as content.
Neighbourhood Pages Are the Real Asset
A national portal can generate a page for every area and cannot know any of them. You can. That asymmetry is the whole opportunity, and it is why neighbourhood content outperforms both city-level pages and listing pages for a local agent.
The difference between a neighbourhood page that works and one that does not is whether a local reads it and recognises the place. Generated demographic tables and a school list do not clear that bar, because they are exactly what a portal already publishes. What clears it is the things you know from working there: which streets are quieter than they look, what the parking is genuinely like, which period of housing has the recurring problem, what people move here for and what surprises them.
Word counts get recommended for these pages and miss the point. The test is whether the page contains something a competitor could not write without living there. If it does, length takes care of itself; if it does not, more words make it longer rather than better. That is the data density argument from programmatic SEO, applied to a category where the temptation to generate at scale is strong.
Handling Properties That Sell
Every sale creates a dead page, and doing nothing turns your site into an accumulating field of 404s at exactly the rate you succeed. Decide one policy and apply it automatically, because this is a volume problem no one will handle by hand.
- Redirect to the nearest useful page, not the homepage.
The area or search page covering that property type. Someone who searched a specific address wants that area; the homepage answers nothing and reads as a dead end.
- Or keep the page and mark it sold.
Clearly labelled, with similar current properties alongside. This is often better for exclusive instructions, where the page had genuine content and can become evidence of what you sell.
- Never leave silent 404s at volume.
Occasional 404s are normal. Hundreds of them, arriving predictably as inventory turns over, is a pattern that makes the site look abandoned in the sections that matter.
- Keep sold data as market evidence.
Aggregate what you sold, at what price, and how quickly, and publish it as a market report. That converts expiring inventory into a permanent asset nobody else has, per publishing original research.
Where Agents Actually Get Found
The instruction, the valuation request, and the recommendation rarely come from a listing search. They come from someone deciding which agent to trust in an area, which is a local and reputational question rather than an inventory one.
That means the profile, the reviews, and the named agent matter more than the property database. A seller choosing between agents is reading reviews describing sales like theirs, checking whether you actually work in their area, and looking you up by name. Those are the surfaces to make excellent, and they are covered in local businesses in AI search and getting more reviews.
It also means the questions surrounding a transaction are more valuable targets than the transaction itself. What sellers should fix before listing, how long things actually take in your market, what the costs are locally: these are searched by people who will need an agent shortly, and they are questions a portal answers generically and you can answer specifically. The structure for those pages is in service page SEO.
Questions Agents Ask About SEO
- Why do my listing pages never rank?
Because the same listing data appears on hundreds of agent sites fed by the same source, and there is nothing on your version to prefer. Add short lifespans and frequent status changes, and you have thin near-duplicate pages that expire. Portals with far more standing usually win them, which is a structural disadvantage rather than a technical fault.
- Should I noindex my IDX listing pages?
Frequently yes for the low-value ones, though it depends on volume. If listing pages are thin duplicates that expire in weeks, keeping thousands of them indexable spends crawling on pages that will not rank and dilutes what your site appears to be about. Keeping the search interface usable for visitors while limiting what is indexable is a reasonable middle position.
- What should I write instead of chasing listings?
Neighbourhood and market content, which is the one thing a national portal cannot replicate credibly. Genuine local knowledge, current market observations, what it is actually like to live somewhere, and the questions buyers ask you repeatedly. That is first-hand experience, which is exactly what generic property content lacks.
- What happens when a property sells?
Do not let it 404 silently. Redirect it to the most relevant area or search page so any accumulated value is retained and visitors land somewhere useful, or keep the page and mark it clearly as sold with alternatives. Sites that let expired listings disappear accumulate dead ends at exactly the rate they sell property.
- Is it worth competing with the big property portals?
Not on listing searches, and you do not need to. The portals win the inventory query; the local agent can win the questions that come before and after it, which are the ones that produce an instruction rather than a viewing request. Competing where you have an advantage beats competing where the outcome is already settled.
