Thoughts from the CTO: AI-based Financial Management

Thoughts from the CTO: AI-based Financial Management

I think I’m starting to like blogging. Today I’d like to look at the history of Personal Financial Management (PFM) and where it’s headed today and into the future.

PFM software has a long history starting in the 80’s with Quicken for the PC. Two decades later, PFM moved to the web with Mint, and it has been an active area of innovation and development during the fintech hype of the last years.

PFM has focused on account aggregation in order to give customers and advisors a comprehensive view of balances and transactions across different types of accounts, assets and liabilities. PFM helps users know their own net worth, monitor their spending and cash flow, better budget, and follow the performance of their investment portfolios. PFM has also been criticized for poor categorization of income and expenses, poor classification of assets and liabilities, and for requiring too much manual work from the users in order to get one’s financial picture truly accurate and well organized.

Artificial Intelligence (AI) and Machine Learning (ML) are helping fintech firms build a new generation of smarter Household Financial Management (HFM) services. AI-based HFM has the potential to fulfill the promise that previous PFM failed to deliver. These services are first and foremost data driven. HFM software learns from data about the customers, their families, their spending patterns, their preferences and their needs. It also learns from user interaction with the system. User actions are interpreted as implicit feedback that the system can use to correct its mistakes and reinforce good decisions.

The ultimate goal (and challenge) of AI-based HFM is financial wellness: helping customers organize their financial lives, giving customers insights about their households’ financial behavior, and making predictions and recommendations in order to help customers make smarter financial decisions. The software is expected to automatically identify investment opportunities and other financial planning opportunities. It should automatically notify the user about changes in income or expenses, overspending or missing payments. It should detect inefficiencies and suggest how to save on costs or adjust spending behavior. Within limits set by the user, it should also be able to move money, make payments and execute smart actions on behalf of the user.

I expect we’ll start to see products that go beyond account aggregation, budgeting and investment performance reporting, and start providing more complex financial wellness services.

Let me know about the fintech topics on your mind – I’m interested and open to your suggestions on what else I can talk (or write) about.

If you like this article, read the last post from my “Thoughts from the CTO Series” about Open Banking.