How to Spot Stale and Relisted Bangkok Condo Listings

Thai property portals will tell you a lot about a listing: the size, the floor, the asking price, sometimes the agent's phone number. What they will not reliably tell you is how old it is. The "posted" date resets when an agent edits a photo. The "updated" date resets when nothing meaningful has changed. And nothing at all tells you that the unit you are looking at has been sitting, unsold, across four different portals since early last year.

This matters because listing age is one of the strongest negotiation signals available in a market with no public sold-price registry. A unit that listed last week at a fair price is a different transaction than a unit that has sat for fourteen months at a fantasy price, even if both look identical on the portal today. The stale one has a motivated seller, or an absent one, and either way you should not be negotiating from the printed ask.

This is a field guide to the four most common patterns, and how nightly price history exposes each one.

Why Bangkok has so many stale listings

Three structural facts produce the problem.

Listing is cheap and delisting is optional. Posting a condo on a Thai portal costs little or nothing, so agents post everywhere and rarely clean up. A unit that sold six months ago through one channel can stay live on three others indefinitely. Nobody's job depends on taking it down.

There is no sold-price registry to force reality onto prices. As covered in our piece on asking versus sold prices, every number on a Thai portal is an asking price. A seller who anchored high faces no public record showing that comparable units sold for less. So the fantasy ask can sit for a year without embarrassment.

Portals compete, so no one dedupes. Each of the eleven portals Vurel reads (LivingInsider, PropertyHub, DDproperty, Hipflat, KaiBaanThai, Kaidee, RentHub, Teedin108, Thailand-Property, Dot Property, FazWaz) shows only its own inventory. An agent blanketing five of them creates five separate listings that no portal will ever cross-reference for you.

The result: a meaningful share of what you scroll past on any portal is not really for sale at the printed price, or not really for sale at all.

Signal 1: the same unit on four portals at four different prices

Agents multi-post. That is normal and mostly harmless. What is informative is when the copies disagree.

Say you find a one-bedroom in a Sukhumvit building asking 8.9M on one portal. Search the building name on the others and you find what is clearly the same unit, same floor, same photo set, at higher asks on two other portals and 8.9M on a third. That spread is not noise. It is a fossil record. The highest posting is the oldest, from when the seller was optimistic. The 8.9M postings are the current position. The seller has already conceded, in public, just not in one place.

Vurel does not merge duplicate listings across portals, and we say so plainly on our methodology page. Building-level matching without a shared unit identifier would be guesswork, and confidently wrong merges are worse than honest duplicates. But not merging has an upside for exactly this use case: when you search a building on Vurel, all the copies show up side by side with their individual asking-price histories. The disagreement between them, which no single portal can show you, is visible in one query.

Practical read: the lowest current ask across the copies is the seller's real public position. The highest stale copy tells you where they started. Negotiate against the trajectory, not the number.

Signal 2: photos that date the listing

Portals recycle "posted" dates, but photos are harder to fake fresh. Things to look for:

• A wall or desk calendar in frame. Agents shoot fast and rarely stage. A 2023 calendar in a "posted this week" listing tells you the real vintage.
• The view from the balcony. Bangkok builds fast. If the photo shows a construction site where a completed tower now stands, the photo, and probably the listing, predates the tower.
• Seasonal tells: Chinese New Year decorations, Songkran mess, Christmas trees in a listing surfaced in August.
• Furniture and appliance styles, watermarks from agencies that no longer exist, or photo dimensions typical of much older phones.
• Identical photos across portals with different furniture in one copy: the older copy shows the previous tenant's setup.

None of this is decisive alone. A dated photo can sit on a genuinely fresh listing because the agent reused an old shoot. But combined with a frozen price or a multi-portal spread, dated photos move a listing firmly into the stale column.

Signal 3: an asking price that has not moved in months

A listing's price behavior over time is more informative than its price today. Vurel has tracked asking-price history nightly since March 2026, and the histories sort listings into three rough types:

Cut and cut again. The seller listed high, the market said no, they adjusted. Two or more cuts means an engaged, motivated seller who has revealed their direction of travel. This is the best negotiation setup available.

Fresh and untested. Listed within the last month or two, no history yet. The ask might be fair or might be fantasy; there is not enough signal to say.

Frozen. No price change for five or six months while sitting unsold. This is the stale case, and it splits two ways. Either the seller is not actually motivated (testing the market, in no hurry, emotionally anchored) or the listing is abandoned and the unit may not even be available. Both are reasons to discount the printed ask heavily or move on.

The frozen case is invisible on the portals themselves, because a portal shows you today's price with no memory. It only becomes visible when someone reads the listing every night and keeps the record. That is the entire point of the history.

Signal 4: the delist-relist game

The sharpest version of the problem is deliberate. An agent whose listing has gone stale deletes it and reposts it as new. The portal shows "posted 2 days ago," the listing jumps back to the top of the search results, and the fourteen months of sitting are erased from public view.

There is no honest way for a buyer to catch this on the portal itself. There are two ways to catch it off the portal.

First, the photos. A reposted listing almost always reuses the old shoot. If the "new" listing's photos carry any of the vintage tells from Signal 2, treat the posted date as fiction.

Second, the history. Because Vurel reads all 2.2 million listings nightly, every listing carries a record of when it first appeared in our read and every asking price since. A listing the portal says is two days old, which our nightly read first saw in April, is exposed by that gap alone. A caveat we owe you: when an agent deletes a listing and reposts it under a genuinely new listing ID, the repost starts a fresh history on our side too, because we do not merge or match listings across time any more than we do across portals. The repost's history begins at zero. What survives is the corpse of the old listing, still searchable by building, with its full price record. The pair of them, an old dead listing and a new identical one, is itself the tell.

The baseline: what asking prices look like right now

Every stale-listing judgment needs a reference point, and the reference point is the zone median. These are median asking prices, not sale prices, from Vurel's nightly read, August 2026:

Zone	Listings tracked	Median sale ask (THB)	Median rent ask (THB/mo)	Median sale ask per sqm (THB)
Sukhumvit	278,434	8.9M	35,000	154,286
Rama 9 and Ratchada	109,795	4.29M	21,000	102,083
Sathorn and Silom	66,483	9.18M	32,000	162,162
Khlong Toei and Rama 4	44,643	9.3M	35,000	175,000
On Nut and Bearing	48,134	4.0M	20,000	90,528
Ladprao	37,220	4.0M	19,000	75,242
Bang Na and Lasalle	41,401	4.95M	18,500	62,153
Nonthaburi	48,713	2.999M	12,000	38,750

Zone-level pages with the full set are at /area.

The use is simple. A frozen listing asking 20 to 30 percent above its zone's per-sqm median, with no cuts on record, is a seller who has not engaged with the market in the entire window we can observe. The median is what the market of current sellers is asking; a stale outlier above it has been rejected by every buyer who scrolled past for months. Either open far below the ask or spend your time on a listing whose history shows a live seller.

A working checklist

Before contacting an agent about any Bangkok condo listing:

1. Search the building name across portals, or in one query on Vurel. Note every copy of the unit and each copy's current ask. The spread is the seller's public concession record.
2. Scan the photos for date tells: calendars, balcony views, seasons, watermarks.
3. Check the asking-price history. Cuts are good news for you. A flat line over many months is a warning.
4. Compare the ask against the zone's per-sqm median asking price. Ask what justifies any premium.
5. If a listing looks brand new but its photos or a dead twin listing say otherwise, negotiate as if it has been sitting the whole time. It has.

What this method cannot tell you

Honest limits, because the method is only useful if you know its edges.

Asking-price history starts in March 2026. A listing that has sat since 2024 shows history only from the point our nightly read began; its earlier life is invisible to us just as it is to you.

We do not merge duplicates across portals or across delete-and-repost cycles. You get all the copies and all the histories; connecting them to one physical unit is judgment work that stays with you, usually via photos and floor plans.

None of this produces a sold price. Thailand has no public record of what units actually transact for, so every figure here, medians included, is an asking price and should be read as a ceiling, not a value.

And posted contact details exist on more than half of listings, about 56 percent, but the distribution is bimodal: some portals post contacts on nearly every listing, while DDproperty and Hipflat post almost none. A stale listing found on a contact-free portal may have a twin with a phone number on another.

---

Vurel reads 2.2 million listings from eleven Thai portals every night, across every zone, and keeps each listing's asking-price history from March 2026 onward. We do not merge duplicates, and we think that honesty is worth more than a tidy count. See how a real listing's history looks at /sample, or read exactly how the data is collected at /methodology.