12 Sep 2023John Macdonald

AI can be the new ‘killer’ tech in maximizing the return on your NPL real estate portfolio

We all know when a loan becomes non-performing, and the lender loses faith the borrower will not honour the debt, it is then in the bank’s interest to keep the value and number of NPLs to an absolute minimum - never good to have them on the balance sheet!

Over the last 10 years, according to PwC data, NPLs with a face value of around €700bn have traded as portfolio transactions.

When you are tasked with overseeing and handling large numbers of NPLs, a first quick assessment needs to be made of each property to determine the best strategy – sell the asset at a discount to recover some asset value quickly, hold for a better outcome, or another route, such as to refurbish and then sell.

A case could also be made for holding the asset and to generate some rental income, or selling on the NPL portfolio on the open market as it will take your organisation too much effort to remedy.

But there will always be challenges and frustrations. One is surrounding data integrity issues. This is where you are unable to make a quick property assessment as your portfolio research uncovers a multitude of data quality issues, inconsistencies, data gaps, and you are left having to resolve any conflicting information.

Resolving gaps and data anomalies can be frustrating because it takes data analysts hours of painstakingly manual work that will slow down asset value recovery. Typically, a bank would either assign analysts and asset managers, or outsource data remediation work to a credit servicing company at great expense to handle their data issues as well as their loan workouts.

A further challenge is being unable to determine a strategy. This can come from difficulty in planning a portfolio strategy by having data that is not accurate and reliable which impacts on your forecasting.

Then there is being under-informed about the portfolio value, and with hindsight, wishing you were more informed about the assets within the NPL portfolio at the time when you were bidding on it.

There are solutions, however.

What AI can do is deliver significant time and cost savings. Through AI, your property data is gathered automatically from a wide range of public and private sources and, with machine learning, standardises and enriches it to offer you clean and accurate and usable data.

By using AI, the company could have radically increased understanding and accounted for the expected amount of final manual remediation work needed, resulting in more time for decision making and potentially adjusting their bid price accordingly.

It can also reduce your reliance on, or else eliminate the need for, debt servicers altogether.

The benefits are significant. Here I’ve outlined just five immediate ones:

• An improved asset recovery rate. Having a fast and clear way to ingest, analyse and remediate your real estate data improving data quality and speeding up the asset recovery process.

• Provide your third-party debt servicer with better quality data which means they can concentrate on fixing the underlying asset management problems instead of cross referencing, validating and fixing data gaps or anomalies.

• Better data gives you greater insights into your existing portfolios and offers higher accuracy which gives you better predictions and forecasting capability, and the time you need to think strategically.

• Lower operational costs and your teams can be focused on areas where they can add the most value.

• Being able to put in competitive / realistic bid prices for new portfolios you are looking to acquire by sampling them first using the AI software to assess their overall data quality.

So why does AI matter for you? 

Firstly, you can work with clean and enriched data, enabling more accurate insights into each portfolio, which in turn gives you better predictions on which to base your strategies.

Secondly, with accurate information, your team can save substantially on time, releasing capacity to work on other portfolios, or take on higher value work.

Ultimately, you can switch focus from micromanaging the details, to focusing on meeting your overall business objectives.