If AI could do one thing for your business, what would it be?

There is no shortage of conversation about artificial intelligence (AI) in the mortgage market at the moment, but much of it feels either overly technical or driven by what systems providers want to sell rather than what advisers actually need.

Over the next three months, I am going to use this column to take a step back and look at AI from a broker’s point of view, starting with a simple question: if AI could do just one thing for your business tomorrow, what would you want it to be?

Moving beyond the noise
It is easy to get caught up in the wider narrative around AI, particularly when it is presented as something that will either transform everything overnight or replace large parts of what advisers currently do. In reality, most firms are not there yet, and many are still trying to work out where it fits, what is worth paying for, and whether it is something they should act on now or wait and see how it develops.

That uncertainty is understandable, because the market is full of partial solutions. Many systems now claim to have AI built in, but very few feel complete, and even fewer clearly solve a defined problem. As a result, there is a risk advisers either ignore it altogether or invest time and money without a clear outcome in mind.

Starting with the right question
Rather than beginning with the technology itself, it is far more useful to start with your business. Where does time go in your firm today, what parts of the process feel repetitive, and where are the pressure points that slow cases down or create frustration for clients?

For some, the answer will sit firmly in administration. Document collection, chasing clients for information, checking payslips and bank statements, and making sure everything is in the right format can take up a significant amount of time. These are all areas where AI, particularly when combined with client-facing portals, has the potential to remove friction and free up resource.

For others, the issue may be communication. Clients increasingly expect regular updates, quick responses and a level of availability that is difficult to maintain within normal working hours. AI-led tools could help bridge that gap, handling straightforward queries, providing status updates, and ensuring clients feel informed even when an adviser is not immediately available.

There is also the question of lead management. Being able to identify which enquiries are most likely to convert, and where to focus time and effort, could change how firms approach new business, particularly in busier periods.

Starting with the right question
Rather than beginning with the technology itself, it is far more useful to start with your business. Where does time go in your firm today, what parts of the process feel repetitive, and where are the pressure points that slow cases down or create frustration for clients?

For some, the answer will sit firmly in administration. Document collection, chasing clients for information, checking payslips and bank statements, and making sure everything is in the right format can take up a significant amount of time. These are all areas where AI, particularly when combined with client-facing portals, has the potential to remove friction and free up resource.

For others, the issue may be communication. Clients increasingly expect regular updates, quick responses and a level of availability that is difficult to maintain within normal working hours. AI-led tools could help bridge that gap, handling straightforward queries, providing status updates, and ensuring clients feel informed even when an adviser is not immediately available.

There is also the question of lead management. Being able to identify which enquiries are most likely to convert, and where to focus time and effort, could change how firms approach new business, particularly in busier periods.

Keeping it practical
Looking ahead, there are bigger questions about how the role of the broker may evolve, and that is something I will come on to in the next article. There is also a wider debate around how lenders are using AI/wider technology and what that might mean for distribution over time.

For now, however, the focus should remain practical. AI is not a single solution that can be applied across the board, and it is unlikely to be adopted in one step. It will come through a series of smaller changes, each addressing a specific part of the process.

A question worth answering
So before looking at platforms, systems or providers, it is worth asking a much simpler question. If one part of your business could be made easier, quicker or more consistent tomorrow, where would you start? Because until that question is answered, it is very difficult to know what role AI should play at all.