Blog

4 strategies for success in AI

Financial Institution: AI

Business man looking out window

Artificial intelligence (AI) is poised to revolutionize the banking industry. Through the power of sophisticated algorithms and machine learning, experts predict AI will create $1 trillion in savings for the financial industry over the next 10 to 15 years. Banks that implement these technologies can realize a 22% reduction in operating costs coupled with a 34% increase in revenue growth. It’s clear that banks that implement AI will gain major competitive advantages, while those that do not risk getting left behind. However, mid-sized banks might find the task daunting. You can position your bank for success with the following four strategies. 

1. Make AI cross-functional

Where an AI project resides in the organization can make or break its effectiveness, so it’s important to ensure artificially intelligent technology is cross-functional. In other words, solutions should be designed to benefit all relevant departments and stakeholders. 

For that reason, AI projects should exist outside of traditional IT so all departments have equal input. Early adopters in various industries recommend creating a Center of Excellence (CoE) to centralize AI initiatives. The CoE can be comprised of representatives from multiple departments who collaborate on implementation, similar to a shared services model where:  

Staff establish project vision

Staff members identify use cases for AI within the bank and outline desired outcomes. For example, the bank might want to implement integrated receivables that automatically match payments with remittance data. The result is increased operational efficiency that saves money, enables staff to focus on high-value activities and creates a compelling new product that grows revenue. 

It’s important to get input from everyone who will be affected by your project. Doing so will help you identify potential challenges and roadblocks and additional opportunities to increase efficiency. Not only that, but it will help you secure buy-in bank-wide, so that all staff members have a hand in development. That spirit of collaboration will discourage unwarranted pushback and foster a smooth transition to your AI solution.

The vision is embedded bank-wide 

The CoE develops a winning solution that achieves the desired outcomes. Its members identify new roles and processes required within various departments and harmonizes interdepartmental efforts. It is responsible for establishing a shared vision with clear expectations, which is key to project success. 

Your CoE will also collaborate with external resources during the design and develop phase. Think of your CoE as the central AI hub: all projects will be managed by its members, who will work with staff and outside partners to ensure project success. 

This centralized CoE approach is used by some of the largest banks, including J.P. Morgan Chase and Deutsche Bank, but it also scales for mid-sized banks and credit unions that must rely on external resources to successfully implement artificially intelligent solutions. 

The application is built

Internal and external resources collaborate to build an AI-driven application that meets project objectives. Many banks work with third-party partners who understand the needs of financial institutions and have extensive experience implementing artificially intelligent solutions for banks. 

It's important to choose the right partner for your bank. Experience and expertise are crucial, but you also need to assess potential partners for cultural fit. When your values and personalities mesh, your projects stand a much better chance at achieving desired outcomes. 

Success stories are shared bank-wide 

Successes are shared across the organization to demonstrate how AI can improve operational efficiency, save money and propel revenue growth. Success stories not only prove the value of AI to leadership, they can spark inspiration that leads to new successes. 

Commit to following up with project participants to identify and share successes. You might send emails bank-wide, laud staff members during meetings or even hold small parties to celebrate your triumphs. 

2. Always start with business needs

Successful AI projects prioritize business and end-user needs. Project leaders understand that while data drives AI projects, it can also hold them back.  

It would be a mistake to attempt to harmonize every aspect of your data before you move forward with a project. Instead, identify smaller segments in which AI can make an immediate impact, then develop solutions around that. 

Recognize that “perfect” doesn’t exist. Artificial intelligence is budding technology, and perfectionism derails projects. This doesn’t mean you can’t build a chatbot that delivers personalized offers or implement robust fraud detection; only that you should begin in one area and branch out from there.  

For example, let’s say your CoE envisions a virtual assistant application that employs natural language processing (NLP) to hold conversations with customers. In your vision, customers can verbally direct the application to check their balance, transfer funds, find the nearest ATM or change their access codes.

It’s an ambitious and worthwhile project, but tackling every aspect of the data at once will slow development and increase costs. Instead, prioritize the most important features and build those out first. Perhaps you’ll start with a chatbot interface that allows customers to interact via text. Once that concept is proven, you can move on to implement the verbal conversation features. 

Ask yourself, your team and your customers: what are the most critical functions that need to be improved? Which improvements will yield desired outcomes such as improved efficiency, enhanced customer experience and revenue growth? If you make a list of ten ways AI can help your bank, which single item grants the greatest benefits?

By establishing priority, you can address data architecture and quality issues incrementally, while simultaneously making progress on AI. Over time, both your data and AI tools will improve.  

3. Plan for quick wins

Most bank budget cycles span a few months to a year. However, full-blown AI development cycles can span multiple years. That’s because the technology first learns with human supervision, then gradually on its own. If you try to do everything at once, it will be a very long time before your applications are live – if they ever make it.  

Your first AI project should focus on delivering quick wins well within a single budget period. As stated, that means prioritizing use cases that deliver immediate value.

Want to improve the customer experience? Start with account self-service, then move on to chatbots, virtual assistants and biometrics/facial recognition. Want a robo-advisor that can give customers sound financial advice? Start with a program that can deliver personalized offers to small business customers. Want to use non-traditional data to inform credit and underwriting decisions? Start by automating Know Your Customer compliance.  

Banks that successfully implement AI start by launching pilot projects comprised of minimum viable products – that that have just enough functionality to demonstrate their full potential. MPS help you get to market faster and minimize development costs.

The MVP pilot approach enables you to troubleshoot issues and complete quality assurance checks. Most importantly, it allows you to prove the value of your application and secure buy-in from senior banking leaders. It’s the quickest, most cost-effective way to convince stakeholders to open their wallets to fund larger initiatives.  

4. Build a digital culture

Artificial intelligence is revolutionary technology that moves your bank into uncharted territory. Like a ship at sea, your entire crew must work together to successfully implement AI. That means it’s not enough to simply establish a CoE or hire a data scientist. The right leadership is critical, and leaders must recognize the need to for new roles, new processes and new ways to collaborate. 

Key leadership roles include: 

  • Translator: Someone who connects business and technical stakeholders. The translator must be well-versed in both the business interests of the bank and the needs of the technical team. This person will serve as the liaison between both parties, ensuring clear communication and fostering collaboration. 

  • Evangelist: This person champions AI projects across the bank. They’re responsible not only for pushing for its adoption, but also for ensuring all stakeholder and departmental needs are met. They talk with department heads, staff, customers and leadership to identify ways AI can help the bank, then work toward its successful adoption.

  • Ethicist: Some companies are hiring digital ethicists who measure the impact of AI on consumers and evaluable potential bias in machine learning. Microsoft is one company that has introduced this role into its organization. If you do not have an ethicist, leadership should at minimum evaluate how any AI initiative will impact customers and work to eliminate potential for discrimination. 

Other important roles include data staff who are charged with managing data ownership, governance, quality and technology. Existing team members might be able to fulfill some of these roles; otherwise, you’ll need to hire new, qualified staff to fulfill AI-related duties. The need for qualified personnel is why experts predict AI will increase employment opportunities in the financial sector. That doesn’t mean increased costs, since the savings afforded by artificial intelligence are vastly greater than the staff investment it requires. 

New roles, responsibilities and processes will cause a cultural shift in your organization predicated on a truly digital, insight-driven environment. Leadership must guide the ship to ensure a smooth transition and that the ongoing needs of all stakeholders are met.  

Don’t be afraid to consult the experts

Navigating your first AI project is a daunting task for any bank, especially mid-sized banks that do not have on-site IT support to “speak the language.” While you might be able to clearly envision desired outcomes, you might struggle to understand how to get there. That’s where partnerships come in. 

External resources such as consultants and technology firms can ease your transition to AI. The best partners understand the needs of banks and the challenges you face, and they have extensive experience developing successful AI projects for financial institutions. 

If you’re just testing the waters, you can seek advice from potential partners. Most will be happy to discuss ways artificial intelligence can benefit your bank. They’ll probably even be able to propose ideas and efficiencies you haven’t thought of. From there, they can map out a plan for your first pilot project, along with expected outcomes. You can use that information to help convince leadership to invest in your pilot.  

Partnerships are ideal for mid-sized banks. They allow you to retain your core focus on customers and financial services, and they complement your strengths with cutting-edge skills and technology. They make it so you don’t need to hire full-time technology staff to implement advanced artificial intelligence into your banking processes. They eliminate the ebb and flow of your bank’s internal staff. 

Partners bring an innovative mindset that can put your FI at the forefront of the AI revolution. They represent an agile approach that speeds development and streamlines processes. Outside resources also make it easy to scale: rapidly jumpstart projects with a resource push, then reduce workload as your project matures. 

You don’t have to go it alone, and you don’t want to. Even the largest banks rely on third-party partnerships to ensure their AI projects are successful. Your selected partner will collaborate with your internal team to bring your vision to life. 

Vision, leadership and execution. These qualities are critical for AI success in banking. Work to make AI cross-functional, prioritize business needs, plan for quick wins and build a digital culture. These four strategies build the foundation for successful AI implementation that delivers immediate value and secures your bank’s future.  

White paper: A Banker’s Official Guide to Artificial Intelligence

Is your organization ready for the AI revolution?

RECOMMENDED RESOURCES

Keep your business moving forward

Smiling woman in business meeting

Blog Financial Institution

Top 4 cases for AI in banking

Discover the top four use cases for artificial intelligence in banking. Banks that get on board now stand to gain a major competitive advantage.