We analyse the entire population of UK property-holding companies, every charge, every late filing, every Gazette notice, and a generation of HM Land Registry transactions. The reports below are the data we publish openly. The models drive the sourcing inside the GalimAI portal.
Coverage limited to England and Wales: HM Land Registry and Companies House data do not extend to Scotland or Northern Ireland.
Three production models drive what shows up in the portal: a sell-timing classifier on the company side, a letter response prediction on the outreach side, and the broader ML-versus-rules architecture that combines them. These are the engineering pieces, written for people who want to know what is actually under the hood.
How the GalimAI sell-timing classifier ranks ~1.97M property-holding companies on probability of disposal in the next 6 to 18 months, the signals it weights, and how to read its outputs in the portal.
Read the methodology → Response PredictionWhat 160,000 real letter outcomes taught us about which owners reply to direct-to-vendor outreach, which features carry signal, and how we score letter recipients before a pack ever goes to print.
Read the methodology → ArchitectureRules can spot a single trigger. Machine learning weighs hundreds of overlapping ones. We explain the hybrid architecture GalimAI uses, where rules win, where ML wins, and where most rules-only sourcing tools stall.
Read the methodology →Companies in formal distress, owners under bridge-finance pressure, dissolutions accelerating, late filings stacking. These are the live reports we publish from the GalimAI dataset for England and Wales.
The current trajectory of distress events across UK property-holding companies, with year-over-year deltas.
The London cut of the distress data: borough concentration and the formal-event subset.
Where distress concentrates by region and what that means for buyer reach.
Year-over-year dissolution rates across property-holding companies and the 24-month signal window.
The cohort of property companies sitting on bridge loans, including 11k to 15k with charges already two years overdue.
How much bridge debt rolls into 2026 and which segments of borrowers face the steepest squeeze.
Multi-charge profiles, gearing thresholds, and the companies most exposed to rate moves.
The signal stack that flags owners showing genuine financial stress versus those that just look stressed on paper.
What Gazette notices, charges, and filing histories together reveal about owner-side legal pressure.
What persistent late filings predict about a property company over the following 24 months.
Why a single signal misleads and which combinations actually predict a sale event.
The portfolio-level view: owners surfacing more than one independent sell signal at the same time.
Who owns UK property at the company level, how long they have held it, and which cohorts behave most like motivated sellers.
The cohort of UK property-holding companies that have held the same freehold since 2003 or earlier, plus the 38k to 45k with a 65-plus director.
The demographic slice that disproportionately maps to estate-driven and retirement-driven sales.
The PSC and director overlaps that flag family-held SPVs, and what they imply for sale catalysts.
Origin of ownership (inheritance, development exit, refinance-rollover) is one of the strongest single predictors of disposal behaviour.
The unencumbered cohort, why they sell less often, and which subgroups do still move.
Short-tenure SPV behaviour, the 24-month dissolution curve, and what that reveals about buy-and-flip activity.
Which buy-to-let segments are exiting, the regional mix, and the financing footprint they leave behind.
What rising operating costs and rate normalisation mean for portfolio gross yields and exit timing.
Bridging, second-charge, development finance, auction finance, portfolio lending. The cohort-level reports specialist lenders use to size pockets of demand and stress.
Who actually carries bridging debt across the UK property-company population, by lender concentration and loan vintage.
A profile of the active 2026 borrower base for business-purpose bridging: sector mix, size bands, and refinance windows.
The slice of commercial bridging where charges have run past term, where refinancing demand really concentrates.
The second-charge population, sizing, and how to find borrowers before the broker channel does.
Which UK property companies are statistically most likely to sell at auction versus on the open market.
Portfolio-scale landlords (10 or more charges) and the refinance churn pattern that defines them.
What development-finance maturity and charge-pattern data tell us about which developers are coming off project.
Where formal distress events sit against the lending population, with the segments showing the steepest concentration.
How specialist lenders can use Companies House signals to find on-pattern borrowers earlier in the cycle.
Who the active UK buyers actually are, where they are, how fast they move, and where the demand stack has thinned.
Newly registered property-buying entities formed in 2025 and where they concentrate.
The active-buyer population in the last 24 months by region and acquisition cadence.
Regional intensity of buyer activity, mapped against transaction volume.
Buyers with the highest acquisition cadence in the dataset and the asset patterns they favour.
How buyer demand splits across residential, commercial, land, and mixed-use in the current cycle.
Where deals are failing on the buyer side and which buyer signals predict completion.
The overlap between agent listings and where genuinely off-market opportunity actually sits.
London Gazette notices, the buyer-window they open, and how they slot into the wider distress stack.
The notice taxonomy: receiverships, dissolutions, charges, and the distinct windows each one opens.
How long the actionable buyer window stays open after a Gazette notice publishes, by notice type.
Why fall-throughs are running where they are, and which seller cohorts they map to.
The structural and cyclical drivers behind weak completion rates in the current market.
The operational playbooks behind the data. How to use the signals once they surface, and the sequencing that makes them convert.
End-to-end methodology for sourcing off-market deals from Companies House and HM Land Registry signals.
The signal taxonomy: catalyst signals, friction signals, and the combinations that genuinely flag motivation.
The temporal patterns that anchor the sell-timing model and where the signal density peaks.
What letter response data tells us about timing, sequencing, and recency relative to signals.
Letter structure, tone, and the variants that genuinely drive replies in the UK property context.
How to separate buyer intent signals from noise on the demand side of the market.
The constraints and catalysts specific to selling a property with a sitting tenant.
What BMV really means in 2026, how to verify it, and where the genuine BMV cohort sits in the data.
City-level breakdowns of off-market opportunity and seller density across the major UK markets.
Each city page distils the underlying GalimAI dataset down to a single market, with the seller signals, ownership patterns, and finance footprint specific to that geography.
Coverage limited to England and Wales: HM Land Registry and Companies House data do not extend to Scotland or Northern Ireland.
The companion set: sell-side conditions, friction, and active-buyer depth by city, for owners weighing a faster off-market exit.
The reports here surface the patterns. The portal lets you query the underlying companies, owners, and signals directly.
Try the portal Read case studies Browse the full blog