Listing pages versus brand pages on agency sites

A listing page sells one property to a buyer and has a shelf life. A brand page sells your agency to the person deciding who to instruct, and has none. A boutique agency needs both, built differently. Most sites collapse them into one feed of properties and then wonder why sellers never call.

What is the difference between a listing page and a brand page?

The difference is ownership and lifespan. A listing page describes an asset you are temporarily responsible for. A brand page describes the agency, which is the only thing on the site that is permanently yours.

Listing pages arrive from a CRM or a portal feed. They are transactional, high intent, keyed to an address, a street, a postcode, a price bracket. They also disappear. A property under offer in March is dead weight by June unless you decide otherwise.

Brand pages are the homepage, the page explaining how you work, the sell-with-us page, agent biographies, the sold archive, one or two area pages, and the journal if you will actually maintain it. They accumulate links, references and trust. They are what an AI answer engine has to read when someone asks which agency handles canal houses in the Grachtengordel.

The asymmetry is that listing pages feel free, because the feed produces them, so agencies pour everything into the property grid and leave the brand side to a single paragraph on an about page. That paragraph is then the entire case for a seven-figure instruction.

Which pages win the instruction, and which win the buyer?

Buyers land on listing pages. Sellers land on brand pages, then read your listings as evidence.

This is the part that gets missed. A seller comparing three agencies does not only read the sell-with-us page. They open your current properties and look at how a home like theirs was photographed, sequenced and written about. Your listings are your portfolio whether you treat them as one or not.

So listing pages do double duty and brand pages do not. That has a practical consequence: the property description cannot be the 90 words the portal feed supplies. It needs the approach, the rooms in the order you would walk them, the plans, the neighbourhood and the facts, written by someone who has stood in the building. If your listing text reads identically on Funda, Rightmove and your own site, your own site is adding nothing to the decision.

The same logic applies in adjacent trades. An architecture practice is judged on the depth of its project pages rather than the count, which is the argument I made in what makes an architecture website win awards.

How should a listing URL be structured so it survives the sale?

Use a readable, permanent path built from the address, not from a database id or a status. Something like /properties/keizersgracht-212-amsterdam rather than /property.php?id=48127.

Google's own guidance on URL structure asks for simple, descriptive, stable URLs. The stability matters more here than in most sectors, because a property that sells well gets shared, linked and archived, and that link equity is worth keeping.

Keep the price and the status out of the path. A URL containing /for-sale/ becomes a lie the day contracts exchange, and renaming it throws away everything the page earned.

What happens to the page when the property sells?

There are three options and only one of them is any good.

Delete it, and you serve a 404 to every link and every buyer who bookmarked it. Redirect it to the search page, and you create the classic irrelevant-redirect pattern that search engines treat as a soft 404. Or keep it, mark it sold, and let the photography stay.

I ship the third. The sold archive is the most persuasive brand asset a boutique agency owns, because it is the only proof that your presentation works repeatedly rather than once. Set availability to sold in the structured data, remove the price if the seller's agreement requires it, keep the images and the plans, and add one line at the foot of the page inviting valuations for similar homes on that street. An expired listing then becomes a permanent lead source for a micro-market you already understand.

How many brand pages does a boutique agency actually need?

Fewer than most agencies build, each carrying real weight. My working set is seven.

• The homepage, with one property at full width, your point of view in a paragraph, then the portfolio.
• A page on how you work, including how you price.
• Sell with us, aimed only at the person holding the keys.
• The sold archive, treated as a portfolio rather than a graveyard.
• Agent biographies with real photographs and direct contact details.
• One or two area pages covering places you genuinely dominate.
• Contact.

The temptation is to publish an area page for every town within an hour's drive. That produces a set of near-identical pages with a name swapped, which is the thin-location-page trap builders fall into and which I unpicked in home builder websites that book site visits. One neighbourhood covered with real knowledge, transaction context and photographs will outrank ten generic ones.

What belongs on the how-we-work page that is not on everyone else's?

The things your competitors will not say out loud: what you decline, how you are paid, and how many instructions you take at once.

A boutique agency's entire premise is selectivity. If the site does not state the limit, the premise is invisible. Say that you take a fixed number of instructions per agent, or that you do not list a home you cannot photograph properly, or that you turn down properties outside a defined type. That is a claim a rival with a 4,000-listing feed cannot copy.

Fee transparency does the same work. Publishing a structure, even a range with conditions, filters enquiries before they reach an agent's diary. I made the full argument for showing numbers in what an architecture website costs in 2026, and it transfers directly: the people put off by a published fee were never going to instruct you.

The Modern House in London is the clearest public example of a firm that treats its non-listing pages as the product. Its editorial and its stated criteria for what it will represent do more selling than any individual property page. You can look at what they publish and see the split working.

How do you keep listing pages fast when photography is the product?

By loading one image immediately and the rest only when the buyer reaches them, and by holding to the published Core Web Vitals thresholds rather than a vague sense of fast.

Those thresholds are public: a Largest Contentful Paint under 2.5 seconds, Cumulative Layout Shift under 0.1 and Interaction to Next Paint under 200 milliseconds, measured at the 75th percentile of real visits. Property pages fail the first and second constantly, because a 4 MB hero and an unsized gallery are the two easiest ways to break them.

A rule of thumb I use on property builds, not a measured study: 20 to 30 images per listing, the hero under 250 KB, gallery images under 200 KB each at around 2,000 pixels wide, served as AVIF with a WebP fallback, every image given explicit width and height so nothing shifts. Floor plans get their own zoomable treatment rather than being dropped into the gallery, because on a phone an unzoomable plan is decoration.

Assume the buyer is abroad, on mobile, on a hotel connection. That is a normal condition in this market, not an edge case.

Does structured data on a listing page do anything?

It makes the page legible to machines. It does not currently buy you a rich result for a residential sale in most markets, and anyone promising one is overselling.

Google's structured data gallery has no general residential-sales listing type, so the honest framing is machine legibility rather than search decoration. That still matters more each year, because answer engines parse the markup when they summarise who sells what and where.

On an agency build I ship a RealEstateListing wrapping a Residence, with the postal address, floor size, number of rooms, an Offer carrying price and currency where permitted, and an availability value that flips to sold rather than vanishing. Breadcrumbs on every listing. A single Organization node for the agency, referenced by id from every page, with the agents as Person nodes attached to it. Nothing exotic, but consistent across the whole site, which is the part most feeds get wrong.

So what does the split look like in practice?

Listings syncing from the feed, enriched with copy and photography you control, on permanent URLs that survive the sale into a sold archive. Brand pages that state a position, name the limits and publish the fees, built by hand and not fed by anything.

That is the structure I use for web design for luxury real estate, where a boutique agency site is typically live in four to six weeks once photography and the listing feed are available. The feed keeps the site current. The brand pages are what make a seller pick up the phone.