Olena Gryniuk talks to Erik Brieva, CEO at Strands, about AI strategy and how technology and data are changing the customer experience in banking.
Strands has been creating highly customizable digital money management software for financial institutions since 2004.
Olena Gryniuk: How can banks make the most of AI technology? It requires a good IT team on place, probably, the appropriate corporate culture, availability of the customer data. What is the most important?
Erik Brieva: There is a 450 billion dollar market opportunity for the applications of artificial intelligence in the financial sector. Banks are currently funding AI initiatives within their organizations, most of them oriented to obtain customer insights allowing them to anticipate their customer needs and suggest the next-best actions.
There are four essential ingredients to achieve an AI strategy:
- Relevant Data: It is required to have access to a huge amount of customer data. Trillions of transactions and events happen every day in banking, most of them contain hidden patterns and relationships.
- Use Cases: Understand what are the needs, the desired results to cover those needs, and define the use cases of application.
- Appropriate Algorithms: Identify the appropriate algorithms to create the valid models for the desired use cases.
- People: Good IT team with the talent, experience, and the appropriate business mentality and corporate culture.
However, those ingredients are not sufficient to succeed if they are not effectively and efficiently combined. To make the most of AI technology, banks need to implement an AI platform that is integrated with their core and digital systems.
I have seen many banks trying to develop artificial intelligence software and applications by themselves, mostly unsuccessfully. Even though these banks have direct access to a vast amount of data, identify the use cases, know the algorithms, and have put together a group of data engineers and data scientists, they haven’t come up with a solution that has a big impact on their business. This is why now banks, instead of trying to build their own AI technology, will plug-in an existing and proven platform that provides pre-configured rules and models, and out-of-the-box actionable insights that they can use right away for their existing banking applications and operations. These banks will then be able to develop new algorithms and enhance such a platform, to generate more insights and additional solutions.
OG: Customer data is another topic for discussion here. On the one hand, data is not a problem these days, as banks have many data sources in their hands, and thanks to Open Banking – getting the data became much easier. On the other hand, banks should have the proper infrastructure to support, acquire, and analyze a vast amount of data, comply with all the regulations connected with the usage and sharing of personal data, etc. And are the customers willing to deliver their data to be better served by banks? What would you comment on here?
EB: While there are some concerns regarding privacy regulations, we now know that consumers are willing to share personal data in exchange for personalized offers, to avoid noise and improve their financial wellbeing. At the end of the day, consumers want to feel special, want to be alerted, advised and even amazed by their bank and by other service providers. They crave new experiences. True value comes from understanding what the customer needs and helping them achieve it by presenting them with personalized, actionable notifications and recommendation of next-best actions.
Yet, many banks are still leaning on simple forms of personalization, and therefore failing to engage increasingly demanding and tech-savvy consumers. Bear in mind that studies show that around 75 percent of users aren’t satisfied with the level of personalization they currently receive. This means that banks are failing to create experiences that actually engage customers in any real way. AI-based financial technology generates deep customer insights that are valuable both for the bank and the customers. Ultimately, it helps banks to empower people to better manage their life and business by making financial decisions in a smarter, more transparent and independent way. AI can unleash a substantial increase in revenue, and the key to unlock this potential is none other than personalization.
But, as mentioned above, for banks in order to implement a successful AI strategy, they should rely on third party AI platforms and be able to feed them with reliable and massive data. And that precisely is a limitation for banks, because of regulations that prevent them to share personal or sensitive data. The solution is synthetic data. There are startups focused on generating anonymous insightful data given a sample data set. And they can generate billions of similar data, all the data that is needed to feed the algorithm and create reliable and effective AI models.
OG: Data and algorithmic bias has been the talk of the town at the latest FinTech events. Building awareness about diversity and inclusion is often stated as a priority in many industries, yet many companies fail to set conditions for it, what are your thoughts on this matter?
EB: In the artificial intelligence field, like in many overwhelmingly white and male-dominated tech spaces, efforts to implement diversity policies haven’t been entirely successful. This can lead to unintended consequences and hurt the ethics of AI-powered technologies. Although deploying biased algorithms doesn’t necessarily entail a conscious intention to discriminate against specific groups, developers must act responsibly to ensure their software is as impartial as possible.
The problem with this is that even the most carefully constructed algorithms only have real-world data to pull from. Data scientists define the questions and analyze the data, but the technology will compound the bias in its logic, rendering it dependent on the whims and imperfections of the world. It is crucial that management teams understand the dire consequences that unfair bias can have, both on their business and on society at large.
Data scientists must scrutinize the data-collection methods, look for any biases and work actively to diversify their datasets. Then, given a valid dataset, they can rely on a synthetic data generator to produce massive amounts of data that follows the patterns and relationships of that sample dataset.
OG: In Strands’ Making it personal eBook, you stress on the AI application for personalization of customer approach. Can you give some examples of what can be used by banks in better servicing SME customers?
EB: SMEs are the backbone of the global economy and represent one of the biggest potential sources of revenue for banks, yet they are typically offered banking solutions designed with the retail customer in mind. Our in-house research shows that more than 60 percent of SMEs place cash flow management as one of their top three priorities to help them secure their financial future, but the banking solutions offered today by issuer banks do not address these unique requirements.
At Strands we have been working to provide banks with an integrated platform of digital cash management and commercial payment tools specifically designed to benefit their small-to-medium enterprise customer base. Strands’ Business Financial Management (BFM) solution helps the SMEs to better understand their finances, project their short-term cash flow, provide more liquidity to their business and engage on a different level with their customers. Our solution offers a comprehensive set of tools and insights to empower SME owners to achieve better and more efficient management of their cash flow and working capital needs, manage accounts payables, receivables, budgets and provisions.
It’s incredibly fulfilling to bring the power of AI-enabled solutions to the SME user to allow them to predict income, expenses, forecast balances, receive personalized alerts and notifications and recommend products and services that meet the immediate needs of their business.