How Machine Learning Makes Banking More Personal

      Posted by Nicole Harper on Sep 30, 2016 in Thought Leadership

      The rise of financial technology, or FinTech, has been one of the most exciting developments of recent years. Firms such as Transferwise, Kantox, Monzo and iZettle have proved that tech startups are capable of offering individuals and SMEs quicker, cheaper and more convenient financial services than traditional banks.

      New progress in the area of artificial intelligence and machine learning has created the opportunity for fintech to steal yet another march on banks. Creative application of machine learning has the potential to enable startups to also offer a fairer and more personal service than anything that has come before.

      What is machine learning?

      Machine learning is a subset of artificial intelligence that enables computers to learn without being explicitly programmed. Put simply, computers analyze new information and compare it with existing data to look for patterns, similarities and differences.

      Each time it does this, the machine improves its ability to analyse, predict and classify information, allowing it to make increasingly better data-driven decisions. The already process plays a significant role in areas as diverse as spam filtering, facial recognition, weather predictions and medical diagnosis.

      Fintech services already use machine learning to prevent fraud by flagging unusual transactions and odd spending patterns, and it is far more efficient than manual monitoring. Thus, the smart application of machine learning principles in the field of customer service can ensure fintech stays competitive in the financial space.

      Machines can provide that vital ‘human’ touch

      A common complaint in banking is that low-value customers don’t get the level of customer service high rollers do, and machine learning will enable ‘high-touch’ personal services to be offered as standard to all.

      A current example of this is the use of machine learning software by fintech firms like Kasisto. Here, algorithms are applied to the company's online Q&A offering - the ‘live’ chat rooms where customer service representatives answer questions in real time.

      This enables a ‘human-like chat experience’ where machine learning assists in responding to queries and helps lower the number of actual human representatives needed to run the platform, meaning resources can be allocated elsewhere. The software also analyses the information coming in and uses it to improve its ability to offer a more targeted, streamlined and specialised offering to the customer in the future - something that would require unfeasibly large amounts of staff for a traditional bank.

      Other fintech firms are taking this personalisation a step further. Wallet.AI is a ‘personal finance service’ that uses a person’s financial profile and transaction records to help advise on day-to-day spending. It may warn a user to reconsider spending money on a cab when other cheaper transport options are available nearby or warn how purchasing those new shoes will make it impossible to pay the rent at month’s end.

      Its founder, Omar Green, has spoken about how the tool helps people who otherwise ‘don’t have access to the kinds of financial advice that might make a material difference in their lives’ and could never afford a personal adviser.

      This kind of creative thinking allows fintech to keep pushing the boundaries of what financial services can do and keeps banks on the backfoot.

      All equal in the eyes of the machine?

      Humans are biased beings. Even with the best of intentions, it’s almost impossible to make a decision without an emotional reaction, or to be affected by some preconceived opinion.

      This is a common issue in banking. In Italy, for instance, despite fewer female-owned businesses going bankrupt than male ones and women overall having better credit histories, female entrepreneurs usually find it harder to get finance than male ones.

      Machine learning can help resolve such matters by making decisions based on data, rather than potential gender and cultural biases. Automating such processes also offsets the tendency for banks to focus on bigger customers over smaller ones, helping to democratise the process.

      Kabbage and LendUp are fintech firms already working in this sector. While the human element can still have a role to play, correctly augmenting such process with machine learning is another way to improve personal services, make banking fairer for all and help those who may suffer unfair discrimination.

      WHO WILL lead the way in Fintech’s ‘personalIZATION’ mission?

      As high-street banks continue to decline, increasing personalization will only boost fintech’s reputation as champion of small businesses and the ‘man on the street’. It may seem a paradox, but machines and ‘artificial intelligence’ can personalize banking in a way high-street banks simply cannot.

      Focusing on advancing the fields of machine learning and fintech would be a very intelligent decision indeed.

      -James Baston-Pitt, OnFido




      EDITOR'S NOTE: This article has been reproduced with kind permission from the author. Some minor changes have been made to reflect style considerations. Read the original piece here.

      Nicole Harper
      Nicole Harper

      Nicole is a communications professional with a strong interest in innovation, big data, UX & human-centered design. She brings disruptive ideas to life and tells stories by delivering beautifully crafted and engaging content.

      Get Our Newsletter

      Subscribe to our exclusive weekly newsletter to stay up to date on
      FinTech trends, insights, and analysis