Likely-to-Sell Data: How Estate Agents Predict the Next Instruction

In the third quarter of 2026, 80,077 homes left Rightmove and Zoopla without a sale agreed, a withdrawal rate of 30.6%, according to Propalt's quarterly market data. Each of those owners put their home on the market and has not yet sold it.

That is what useful likely to sell data looks like. It is built from things owners and properties have done in public, such as listing, withdrawing, falling through, lodging an EPC or passing through probate. It is not built from guesses about divorce, illness or redundancy.

So which signals predict the next instruction, what does the evidence say about how long people stay, and how do you build a list without crossing a data protection line?

What does "propensity to sell" mean for an estate agent?

Propensity to sell is the likelihood that a given home will come to market in a given period, estimated from observable signals. A propensity to sell model ranks homes so your letters, calls and canvassing go to the addresses most likely to need you first.

The concession comes first: no data set tells you who will sell. A score narrows a patch of 10,000 homes to a few hundred worth writing to. It does not replace the market appraisal, the local reputation or the follow-up. Treat likely to sell data as a shortlist and it earns its cost.

Reading the signals well starts with knowing why owners behave as they do. Our piece on why understanding vendor behaviour matters to estate agents covers the patterns behind them.

How long do UK homeowners stay before selling?

It depends which group you measure, and the two most quoted figures describe different people.

Measure	Figure	Source
Average time owner occupiers in England have lived in their current home	17.0 years	English Housing Survey 2024 to 2025
Outright owners	23.7 years	English Housing Survey 2024 to 2025
Mortgagors	8.9 years	English Housing Survey 2024 to 2025
Average length of ownership for homes sold in the 18 months to October 2025	9 years, the same for flats, terraced, semi-detached and detached homes	Zoopla, reported October 2025

The 17-year figure describes everyone living in their home today. The 9-year figure describes the people who sold. The gap suggests where instructions come from. Sellers' average of 9 years sits close to the mortgagor figure of 8.9, while outright owners have stayed for more than two decades.

Length of ownership works as a filter. On its own it rarely triggers a sale. A home bought eight to twelve years ago is a reasonable place to look. A home bought eight to twelve years ago that was also listed and withdrawn last spring is a far better one.

To use ownership length as a filter, you need the last sale date attached to the right address. HM Land Registry Price Paid data records the price and date of each sale in England and Wales.

"Length of ownership tells you where to look. Listing behaviour tells you who to write to."

Which likely to sell signals can you lawfully use, and which should you leave alone?

Use signals that come from public records or from the property's own marketing history. Leave alone anything that requires you to infer a person's health, relationships or finances.

Signal	Where it comes from	How to use it
Withdrawn listing	Portal listing history	Strongest signal. The owner has already decided to sell.
Sale fallen through	Listing back on the market after sold subject to contract	Time-sensitive. The owner still wants to move and may be reviewing their agent.
Repeated reductions	Listing price history	Signals a pricing problem the current agent has not solved.
Probate	Grants of representation, searchable on gov.uk for deaths in England and Wales; new records appear about 14 days after issue	Write to the executors, with care over tone and timing.
New EPC	Public EPC register	Sellers' agents must confirm an EPC has been commissioned before marketing, so a fresh EPC on a home with no current listing can precede an instruction or a let. Owners can opt out of public disclosure.
Planning application	Council planning registers	Context. An extension usually means staying; a completed scheme can mean a sale after works.
Landlord selling	Listing and ownership data	Hamptons found 9.2% of homes listed for sale in June 2026 had been rented out in the previous five years, down from 11.3% a year earlier. Fewer landlords are selling, so each one matters more to a lettings and sales agency.
Length of ownership	HM Land Registry Price Paid data	Filter only.

Reductions are the most plentiful of these signals: 83,212 new Rightmove listings nationally in Q3 2026 were reduced below their first listed price (Propalt data).

The signals to leave alone are life events you cannot lawfully observe: divorce, illness, bereavement outside probate, pregnancy, job loss. Health is special category data under Article 9 of UK GDPR. In October 2022 the ICO issued Easylife a £1.35 million GDPR fine, later cut to £250,000 by agreement in March 2023, after it inferred health conditions from customers' purchases and used them for telephone marketing without telling them, which the ICO described as "invisible processing". Inferring a life event from indirect data is the same pattern in a different industry.

What does Propalt's Q3 2026 data show about withdrawn listings?

The withdrawn share rose in every region between Q2 and Q3 2026, and London's share is close to half.

Area	Withdrawn, Q3 2026	Withdrawn share, Q3 2026	Withdrawn share, Q2 2026
National	80,077	30.6%	24.9%
Greater London	14,652	47.3%	39.1%
South West	9,801	33.6%	26.2%
South East	14,957	33.3%	27.1%
East of England	9,203	32.2%	26.3%
East Midlands	5,628	28.3%	24.5%
Wales	3,219	27.6%	22.9%
West Midlands	5,922	27.3%	22.0%
North West	7,173	25.6%	20.4%
Yorkshire	5,469	25.5%	20.7%
North East	2,317	23.9%	18.9%

Two readings matter for prospecting. In London, almost one in two homes that left the portals last quarter did so unsold, so a withdrawn list there is large and needs tight filtering by price and property type. In the North East, the share is lower but homes reached sale agreed in 51 days, the fastest of any region, so a withdrawn home there may signal an owner who expected to sell quickly and did not.

How do you build a likely to sell list ethically?

Start with property signals, contact by post first, and honour every objection. In practice:

1. Build from listing behaviour. Withdrawn, fallen through and reduced homes are signals the owner created by marketing publicly. They are the most defensible place to start.

2. Exclude homes still on the market. Writing to a home that is live with another agent invites complaints and wastes postage.

3. Lead with post. The ICO says postal marketing has no PECR requirement, so consent or legitimate interests are the likely lawful bases. Record a legitimate interests assessment and include a short privacy statement and opt-out route in every letter.

4. Treat phone and email differently. Marketing emails and texts to individuals need consent or the soft opt-in, and marketing calls to numbers on the Telephone Preference Service need consent.

5. Handle probate with care. Address letters to the executors, keep the tone practical, and do not imply you know more than the public record shows.

6. Keep a suppression list. The right to object to direct marketing is absolute. The ICO confirms you can keep the person's details on a suppression list, marked so they are never mailed again.

7. Measure by signal. Track valuations booked per hundred letters for each signal type, so next quarter's budget goes where the instructions came from.

Propalt's Audience Builder builds these lists from audiences including Withdrawn, Sale Fallen Through, Probate and Actively Selling Landlords, and can exclude named agency brands. For the letter itself, see how to write a prospecting letter, and for landlord targeting, how to find new landlords.

📌 KEY TAKEAWAY

Likely to sell data works when it is built from what owners have done in public: withdrawn listings, fall-throughs, reductions, probate grants, new EPCs and landlord sales. Length of ownership is a filter; English Housing Survey data puts the average owner occupier at 17.0 years in their home, while Zoopla found sellers had owned for 9. In Q3 2026, 30.6% of homes leaving the portals did so unsold, rising to 47.3% in London. Contact by post first, keep a suppression list, and never infer life events. Propalt's Audience Builder is built to turn those public signals into a list you can act on.

SOURCES & DATA REFERENCES

1. Propalt (October 2026). Quarterly market data, Q3 and Q2 2026, national and ten regions (get_quarterly_market). Withdrawn count and share (national Q3 2026: 80,077 of 261,667, 30.6%), regional withdrawn shares, North East 51 days to sale agreed. propalt.co.uk

2. Propalt (October 2026). Quarterly market data, national, Q3 2026 (get_quarterly_market): 83,212 new Rightmove listings in the quarter (counted per listing, not deduplicated) reduced below their first listed price.

3. Ministry of Housing, Communities and Local Government (December 2025, updated July 2026). 'English Housing Survey 2024 to 2025: headline findings on demographics and household resilience, Chapter 3.' Owner occupiers 17.0 years in current home; outright owners 23.7; mortgagors 8.9. gov.uk/government/statistics/chapters-for-english-housing-survey-2024-to-2025-headline-findings-on-demographics-and-household-resilience/chapter-3-housing-history-and-future-housing

4. Mortgage Solutions, reporting Zoopla (16 October 2025). 'Sellers of larger homes see bigger gains than flat owners.' Average ownership of 9 years for homes sold in the previous 18 months, across all property types. mortgagesolutions.co.uk/mortgage-news/2025/10/16/sellers-of-larger-homes-see-bigger-gains-than-flat-owners-with-a-122k-profit-zoopla

5. Hamptons (July 2026). 'Landlord sales slow following the Renters' Rights Act.' 9.2% of homes listed for sale in June 2026 previously rented, down from 11.3%. hamptons.co.uk/articles/landlord-sales-slow-following-the-renters-rights-act

6. GOV.UK (accessed October 2026). 'Search probate records for documents.' Records for deaths in England and Wales since 1858; new records online about 14 days after issue. gov.uk/search-will-probate

7. legislation.gov.uk. The Energy Performance of Buildings (England and Wales) Regulations 2012, regulation 7. EPC must be commissioned before marketing. legislation.gov.uk/uksi/2012/3118/regulation/7/made

8. GOV.UK (July 2019). 'Energy performance certificates: opt out of public disclosure.' gov.uk/guidance/energy-performance-certificates-opt-out-of-public-disclosure

9. Lewis Silkin (18 October 2022). '£1.35m ICO fine spells no Easylife for marketing-related profiling and invisible processing.' lewissilkin.com/insights/2022/10/18/1-35m-ico-fine-spells-no-easylife-for-marketing-related-profiling-and-invisibl-102hzc7

10. Information Commissioner's Office (16 March 2023). 'ICO reaches agreement with Easylife Ltd.' GDPR fine reduced to £250,000. ico.org.uk/about-the-ico/media-centre/news-and-blogs/2023/03/ico-reaches-agreement-with-easylife-ltd

11. Information Commissioner's Office (accessed October 2026). 'Sending direct marketing: choosing your lawful basis.' ico.org.uk/for-organisations/direct-marketing-and-privacy-and-electronic-communications/sending-direct-marketing-choosing-your-lawful-basis

12. Information Commissioner's Office (March 2025). 'Right to object.' Absolute right for direct marketing; suppression lists permitted. ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/individual-rights/individual-rights/right-to-object

Contains HM Land Registry data © Crown copyright and database right 2026. This data is licensed under the Open Government Licence v3.0.