New AI tool indicates the likelihood of a property selling – London Wallet

A new tool has been launched that assigns a percentage score indicating the likelihood of a property selling, if instructed by an estate agent.

Called SellScore, the tool’s insights are designed to help agents prioritise vendor enquiries with the highest likelihood of completion, tailor pricing strategies, reduce spend marketing properties unlikely to sell, and improve CRM efficiency through lead scoring, ensuring resources are focused where they matter most.

TwentyEA, which will officially launch the new product at the Kerfuffle conference this week, says the uncomfortable truth for estate agents is that on average only 55% of listings actually sell which means nearly half of all marketing budgets, viewings, and sleepless nights are spent on properties that will never complete.

Built on data and powered by AI, SellScore assesses thousands of metrics. As well as property-specific data points such as potential risk factors, the model also encompasses the market activity specific to each location, such as the supply and demand ratio and likelihood of fall throughs.

After 12 months of testing against real market outcomes, SellScore correctly predicted 86% of all property sales within the top 30% of properties it flagged as most likely to sell.

SellScore gives every property in the United Kingdom a score as to how likely it is to complete if it comes to market. The number of properties in the top three deciles for UK residential addresses are as follows:

In other words, this data snapshot above shows there are in excess of 3.2million UK residential properties which are in the top 10% of all properties most likely to sell if they came to market over the next year.

SellScore is available via online platform, API or secure FTP data transfer. Ultimately, it will help agents improve their lead scoring practices and boost revenue by ensuring that resources are focused on the right listings.

When paired with TwentyEA’s Forecast tool (which predicts which properties will come to market within the next two months), agents get the complete picture: properties that are both likely to list and likely to sell. This is where it becomes a key tool for increasing marketing ROI.

By combining the two datasets, agents can access a hyper-targeted, ultra-high value set of leads. With just 280,168 properties across the UK which feature in the top 10% of both products, these are the hottest prospects for agents – properties that are likely to list in the next two months, but most importantly, will also be most likely to complete.

Katy Billany, executive director of TwentyEA, said: “We’re incredibly excited about the potential of SellScore to help agents prioritise their resource for the properties where it matters most. With only around 55% of listed properties actually selling, the ‘no sale, no fee’ model significantly impacts an agency’s bottom line but with the power of SellScore, agents can decide which fee model is best suited to the listing – up front, pay on completion or somewhere in between.

“SellScore delivers clear, actionable insights which will help agents to explore new fee models and vastly increase their average income per listing. We’ve built this model using detailed historical data from the past four years which draws on highly granular insights. These include property-specific factors such as potential risk factors, location data, and ownership details such as how long the owners have occupied the home and their life stage.

“It also draws on hyperlocal market data such as the demand to supply ratio. When combined with our Forecast tool, agents gain an even more powerful, data-driven advantage to their marketing campaigns, by knowing exactly when the properties most likely to sell will likely come to market.

“SellScore is powered by a vast number of mathematical calculations, refreshed daily through advanced AI, ensuring the model stays accurate and up to date.”

Wembley prepares to host Kerfuffle Conference 2025