Proptech business TwentyEA is making a bold claim about the effectiveness of its use of proptech in identifying which homes in a given area are most likely to come on the market within the next four months.

TwentyEA says it holds the UK’s biggest and richest resource of factual life event data including the largest, most comprehensive source of home mover data compiled from more than 29 billion qualified data points, and is celebrating a five-year milestone for its Forecast tool which uses data algorithms and traditional analysis techniques.

As a result of over half a million pounds being invested in leading-edge technology, the company says that the AI and machine learning used in its forecasting have improved significantly year on year,

Over the five-year period there has been substantial refinement of the AI model, taking it from predicting properties likely to instruct in the next twelve months, to its current position where it is highly accurate in predicting instructions in the next four months.

Data for March 2022 shows that the top 10% of the model successfully predicted 62% of all of the new instructions that came onto the market last month.

TwentyEA Forecast can give every UK property a score as to how likely it is to instruct in the next four months, by looking at hundreds of different data points around the property itself, the people living in it and the local market.

Many of the criteria that are the only selection options in other ‘off market’ tools, have already been built into TwentyEA’s Forecast, such as how long the owner has lived in the property; how old they are; whether the property has been withdrawn from the market in the past; or whether a sale has fallen through, but the property has not gone back up for sale.

This is then combined with hundreds of other data points that might increase their likelihood to move – are they a young couple with a baby living in a small flat, or someone who is recently divorced living alone in a large house with lots of equity?

At the start of this year there were 350,980 properties for sale in the UK, 36% fewer than at the start of 2020, and the lowest amount since TwentyEA began collecting and analysing data in 2008. At the same time, the number of homes available to rent is also at rock bottom, with prospective tenants reporting they have had to jostle their way on to shortlists just to view a property.

Despite many homeowners still wanting to move, a lot are currently holding off listing their property, driven by fear that as there is little on the market, resulting in a struggle to find their next home.

Katy Billany, executive director of TwentyEA, said: “For many estate agents, targeted ‘off-market’ prospecting is a key way of not only finding potential vendors, but also helping clients to find their next home. A double win in terms of getting hold of more stock.

“The beauty of AI and Machine Learning is that by definition they learn over time and so given that our model now has five years’ experience behind it, and that we are now in our fifteenth year of tracking the property market, so it has a great deal of data to learn from, we have the most accurate and effective propensity to move data available to agents.

“Our forecast model has evolved significantly over this time, making it an increasingly more valuable tool for our clients.

“Even a year ago, targeting the top 30% of properties to reach 80% of new instructions was impressive, but now we’ve improved that even further, so that agents only need to communicate with 10% of the properties in their patch to reach 62% of the vendors who will instruct.

This is a remarkable five times better than random targeting. In fact, one of our largest clients generate over 60% of their valuations from our Forecast data.