The rise of Chief Behavioural Officer and how to hack customers’ mind

Financial Services has traditionally been a game of numbers, aggressive product (mis)selling, big bonuses and as a result too many “too big to fail”s. The rise of regulations and technology innovation within the industry has resulted in a course correction. The customer is now getting some attention. And more recently customers’ behaviour is.

Applying behavioural sciences to study how customers make their financial decisions has seen some recent traction. It is critical to bring together cognitive biases and behavioural anomalies and understand how these affect financial decisions. Add to that the impact these financial decisions have on an organisation’s PnL. An executive in a bank capable of doing that would be an ideal fit to as the Chief Behavioural Officer.

The rise of Fintech was accompanied and facilitated by the rise of friendly customer journeys. Today I spend so much, not realising how much money has gone out of my bank account with just a touch. The process is so frictionless that, the act of paying someone is invisible. When it is out of sight, it is out of mind. That’s the simplest way to make people spend more.

That is an example of how an existing process has been made less of a touchpoint. Recently, Merill Lynch conducted an experiment where they asked users to upload their photos. They would run a program to show what the users would look like in 30-40 years. That made the users more conscious of their retirement planning, and shifted their financial behaviour.

Another instance is where the Commonwealth bank of Australia launched an app for customers to set goals. The tool prompts customers to set personalised savings goals and breaks them down to smaller milestones. Since February 2019, users have created more than 250,000 savings goals – 27% of the goals being towards a holiday and 19% towards a property.

Human beings are irrational when it comes to financial decisions. An understanding of behavioural sciences is not just important to win over customers. It is critical to understand the biases that affect existing business decisions that are made within a firm. Leading valuations expert Ashwath Damodaran calls for the need to study these biases as much as the valuation principles used within investment banks.

All valuations are contaminated by bias, because we, as human beings, bring in ourpreconceptions and priors into the valuations. When you are paid to do valuations, that bias multiplies and in some cases, drowns out the purpose of valuation

Professor Ashwath Damodaran

We (my firm Green Shores Capital), recently did an event to identify top financial inclusion firms to invest/track for investments. One of the firms Confirmu, based out of Israel, study the psychological responses from a potential borrower to assess if they were trust worthy or not.

It is one thing going through the borrowers bank account, business plan and reasons for the loan. While all that could be genuine, a borrower might still not have a genuine intention to repay. Especially in countries where there are no credit bureaus, a system to predict the future behaviour of a borrower could be very handy.

While there are several tools to assess the ability to repay, there are very few to assess the intention to repay a loan.

Confirmu’s customer journey takes the users through a chat process, where the customers answer a few basic lifestyle questions, then choose their favourite between a bunch of images, and finally leave a voice answer to a question. Their machine learning powered algorithm gives a rating indicating the customer’s intention to repay.

Banks have started to focus on technology much more than they have ever done. However, as Steve Jobs puts it – ” it’s technology married with liberal arts, married with the humanities, that yields us the results that make our heart sing”


Arunkumar Krishnakumar is a Venture Capital investor at Green Shores Capital focusing on Inclusion and a podcast host.

I have no positions or commercial relationships with the companies or people mentioned. I am not receiving compensation for this post.

Subscribe by email to join the 25,000 other Fintech leaders who read our research daily to stay ahead of the curve. Check out our advisory services (how we pay for this free original research).

5 comments

  1. Hello Arun
    We looked at Confirmu 12 to 18 months ago and at that time it was tough to gauge accuracy of the tool – it needed more time to mature their offering.
    Were you able to see statistically significant impact from using this tool?
    Cheers, Vani

    • I can tell you that they have clients, and have data to demonstrate the tool’s capabilities. Feel free to ping me in LinkedIn if you need specifics.

  2. Great post @Arun.

    Can I be cynical as I wonder how Steve Jobs put in action what he said? `it’s technology married with liberal arts, married with the humanities, that yields us the results that make our heart sing“

  3. On-line lenders (at least the more profitable ones) have been doing this since 2008 and it does not require that the borrower be asked questions or have to choose pictures (which is kind of annoying for customers). You can do it from users’ behavioral, device, social data you collect as part of the loan or insurance application process, along with geo-location and other legally and freely available data, and it has a very significant impact on the default rates.

    See these examples for the before and after results.

    https://zorallabs.com/case-studies/consumer-online-credit

    • I liked the metrics you have captured here. Thanks for that. While you can always tap into everything about a customer from alternate data – to deliver a lending decision, I feel that may not work in societies where data privacy is considered super critical.. In the case of Confirmu, they can plug into the social data as well, but they can work without that too. Its best to have that option. And ofcourse its early days for this field.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.