"Motivated seller" is one of the most over-used phrases in UK property. Everyone claims to find them. Almost no one defines what one actually looks like in public data.
Below is the framework we use at GalimAI to score every UK property owner. Six families of signal. Each rooted in publicly accessible information, scored together, weighted against each other, and used to decide which 1,500 owners get a letter this week.
The point is not "more signals = more motivated." The point is that certain combinations consistently predict a decision to sell within 6–12 months. The combinations are what's interesting.
Signal 1 — Financial pressure
01 Financial pressure
What it looks like: Overdue Companies House filings on a property-holding SPV. Recent change of registered agent. Director resignations. CCJs at the property address. Insolvency notices against the owner or a related company. Active mortgage refinance cycle hitting at a bad time. Property bought at the top of a financing cycle with a mortgage about to reset.
Why it matters: Financial pressure is the strongest single predictor in the dataset. Owners under cashflow stress make decisions faster, accept less, and often want discretion. They will not list publicly because public listing telegraphs distress to the rest of their portfolio's funders.
Signal 2 — Legal events
02 Legal events
What it looks like: Section 21 issued. Eviction proceedings. Court judgements at the property address. Disputes between joint owners. Probate filings indicating recent death of the registered owner. Decisions over a property held in trust where beneficiaries disagree.
Why it matters: Legal events force the owner into a decision window they didn't choose. Probate, for example, typically takes 6–12 months to reach the open market — beneficiaries are usually open to a discreet sale during that period to avoid the cost and time of a full marketing process. Eviction proceedings indicate landlords actively contemplating exit. Court judgements at the address often reveal asset-protection planning that involves selling.
Signal 3 — Ownership structure
03 Ownership structure
What it looks like: Single SPV vs personal name vs trust vs deceased director vs nominee. Recent changes to the People with Significant Control register. Properties owned through entities incorporated specifically for a single asset that's now matured. Companies whose listed business activity diverges from their actual holdings.
Why it matters: Different structures predict different exit patterns. A property held in a personal name by a 65-year-old landlord behaves very differently from the same asset held in an SPV with three director changes in 18 months. Structure tells you the kind of conversation to expect — and whether to offer to buy the property, the company, or both.
Signal 4 — Portfolio profile
04 Portfolio profile
What it looks like: Single-asset owner vs 4-property mini-portfolio vs 30+ landlord. Concentration in a single postcode vs geographic spread. HMO licences vs single lets vs commercial. Average tenure of the holdings.
Why it matters: Different sizes of portfolio behave fundamentally differently. Mini-portfolios are usually held by hands-on operators who often want to consolidate. Large portfolios are professional operations where exit is usually staggered. Single-asset owners are often accidental landlords — high probability of exit when rents stop covering costs. The signal here is not size; it's whether the portfolio size suggests the owner is still actively buying or quietly exiting.
Signal 5 — Timing
05 Timing
What it looks like: How long the owner has held the property. When the last mortgage was taken. When the next EPC certificate expires. When the most recent licensing renewal is due. Whether holdings cluster around predictable life events (retirement, divorce, inheritance).
Why it matters: Property decisions are usually triggered by deadlines, not desire. EPC C is forcing decisions on properties that need £8–15k of upgrades. Mortgage rate resets are forcing decisions on landlords who refinanced cheaply in 2021. Selective licensing renewals are pushing tired landlords out. Timing signals tell you when the conversation will be relevant — and that determines when the letter should land.
Signal 6 — Owner profile
06 Owner profile
What it looks like: Age. Declared business interests. Recent life events visible in public records. Director history across companies. Pattern of past property transactions. Network of business associates.
Why it matters: The owner profile is the difference between "this property has signals" and "this person is ready to talk." A 67-year-old landlord with four HMOs, a recent licence renewal he didn't request, and a clean compliance history is a very different conversation from a 38-year-old fund manager holding the same four HMOs through an active SPV. Same asset. Different person. Different motivation. Different letter.
The example that taught us this works. In September 2024 we sent a letter to a landlord whose properties matched on five of the six signals — particularly his age and the recent renewal cycle on his HMO licences. He replied within two weeks: "At the age of 67, I have decided to hang up my boots, so your letter was quite well timed. I have four HMOs that you may be interested in." He wasn't on the market. He wasn't planning to be. He had simply made the decision he was ready, and was waiting for the right approach.
Why one signal is never enough
The mistake almost every direct-mail-driven sourcer makes is targeting on a single signal. "Probate addresses." "Recent CCJs." "Section 21 issuers."
Each of these works modestly on its own. None of them rises above a 1–2% response rate when used in isolation, because the population that matches a single signal includes a lot of people for whom the signal doesn't actually mean what you think it means.
The 3–6% response rates we consistently see come from combinations: someone who is the right age, holding the right portfolio, in the right ownership structure, with timing pressure visible in the data. Each individual signal is weak. Combined, they're a strong predictor.
What this means for your sourcing
If you're sourcing your own off-market deals in the UK, three concrete takeaways:
- Stop targeting on one signal at a time. Build a scoring model, even a crude one in a spreadsheet, that combines at least three signal families.
- Public records are richer than you think. Companies House, Land Registry, the Insolvency Register, court records, and the People with Significant Control register together cover most of what you need to know. The work is in joining them.
- Test your model on properties that have actually sold off-market. Reverse-engineer five recent off-market sales you know about. Score the owners against your six-signal model. If your model would have flagged them before the sale, it works. If not, recalibrate.
What we built
GalimAI scores every UK property owner against all six signal families, automatically, every week. When a buyer defines their criteria — geography, asset type, portfolio size, owner profile — we surface the owners most likely to be motivated, then run direct-to-vendor letter campaigns under the buyer's brand.
If building this yourself sounds expensive, that's because it is. We've been at it for over three years. The dataset, the scoring model, and the compliance and outreach workflow took us hundreds of campaigns to refine.
Want to see what 50 motivated owners look like for your criteria?
Tell us what you're looking for. We'll come back with 50 owners worth a letter, scored against all six signal families, and a plan to get answers from them.
Book a call Request a sample packFAQ
How do you score the six signals against each other?
Each signal family carries a weight, and within each family individual sub-signals carry their own sub-weights. The weights are recalibrated quarterly against actual response data. Financial pressure and timing tend to carry the highest weights; portfolio profile and owner profile add precision but rarely move the score on their own.
Are these signals legal to use for marketing?
Yes. All the signal families described above are derived from publicly accessible records (Companies House, Land Registry, court records, Insolvency Register, licensing registers, electoral roll where appropriate). We comply with UK GDPR and PECR throughout. The Mail Preference Service is respected; recipients have a clear opt-out path in every letter; data sourcing is documented.
What's the response rate if you target on all six signals?
Targeted across all six families with a handwritten letter, response rates land between 3% and 6% reliably. Untargeted mass-mail to a postcode area produces 0.5–1%. That five-to-twelve-times difference is the value of the scoring model.
Can I see the model myself?
The weights and proprietary scoring logic stay internal. The framework — the six families — is the public part, and we hope it's useful even if you build something yourself. If you'd rather skip the build, book a call.