Platform's tech predicts client exits for advisers

Platform's tech predicts client exits for advisers

The artificial intelligence technology created and used by Praemium can predict when an adviser’s client is set to close their account with 90 per cent accuracy, the platform has claimed.

Mark Sanderson, chief operating officer, told FTAdviser the firm has launched an AI model to its platform which tracks client behaviour and can register when a client shows signs of distress or detect the normal pattern of actions before an account closure.

He said: “A lot of people want the ‘AI’ badge but this really works. Using machine learning, it builds this probability calculator based on how these behaviours imply what a person might do.

“The machine has all the information and simulations and so can register that it looks like someone is going to close their account.”

Using the ‘adviser portal’ part of the platform, advisers can see the insights provided by this AI tool and take action if a client is showing 'signs of distress'.

Mr Sanderson said this allowed advisers to provide value for their clients when they needed them most as the data was available in real time.

He added: “When advisers tell us about the conversations they have with clients off the back of this alert, it is usually always about uncertainty.

“That moment of uncertainty is when an adviser can intervene and have this conversation. It puts the power back in the adviser’s hands to initiate the engagement, get in touch and help.”

He added that in the 10 per cent of times when the system had flagged the concern but it was not an account closure, it was something else such as serious ill health, a divorce or other major life events.

Tim Morris, IFA at Russell & Co, said that although he kept in regulator contact and engaged with his clients, there had been times when clients did not want to continue with an ongoing service.

He said: “It would have been good to get some advance notice and discuss their concerns, and how we could adapt their service accordingly.

“So while I do find it slightly unsettling how closely we are tracked online these days, this is a good example of how this technology can be used in a beneficial manner.”

Ricky Chan, director at IFS Wealth and Pensions, agreed, saying it was “somewhat scary” but that the tool could help advisers become better at helping clients.

He said: “If the AI spots that certain client behaviours, such as unusually more logins to check valuations, means that there’s a high chance of clients leaving or cashing their investments, it may only take a short e-mail or phone call from the adviser to reassure the client.”

However, Mr Chan warned that while having access to this information was helpful, knowing what to do with it and how to prevent clients leaving or making poor financial decisions was “another thing altogether”.

Praemium said the tool had helped flag a number of different client scenarios to their advisers, from a cancer diagnosis to a lottery win.