The transformation to an AI First company using a financial institution as an example.
How enterprises can meet the AI challenge
A new era
Artificial intelligence technologies are increasingly an integral part of our lives, and banks need to deploy these technologies at scale to remain relevant. Success requires a holistic transformation that spans multiple levels of the organization.
For global banking, the consulting company McKinsey estimates that AI technologies could potentially deliver up to $1 trillion in additional value per year.
"Banks need to deploy AI to keep relevant."
to start ?
Many banks are struggling to move from experimenting with selected use cases to scaling AI technologies across the enterprise. Reasons include the lack of a clear strategy for AI, an inflexible and low-investment technology core, fragmented data assets, and outdated operating models that make collaboration between business and technology teams difficult. In addition, several digital engagement trends have accelerated during the COVID-19 pandemic, and large technology companies are looking to enter the financial services sector as their next neighborhood. To successfully compete and thrive, established banks must become "AI-first" institutions and leverage AI technologies as the foundation for new value propositions and distinctive customer experiences.
In this lecture, we propose answers to four questions that can help executives formulate a clear vision and develop a roadmap to become an AI-first bank:
Why do Banks need to become AI-first?
What might the bank of the future look like
What are the barriers preventing banks from deploying AI capabilities at scale?
How can banks transform to become AI-first?
Why does it need an
AI first approach?
Over several decades, banks have continuously adapted the latest technological innovations to redefine the way customers interact with them. In the 1960s, banks introduced ATMs; in the 1970s, electronic card-based payments. The 2000s saw the widespread adoption of 24/7 online banking, followed by the proliferation of mobile "banking on the go" in the 2010s. Few would dispute that we are now in the AI-driven digital age, fueled by falling data storage and processing costs, increasing access and connectivity for all, and rapid advances in AI technologies. These technologies can lead to increased automation and, when deployed after risk control, often improve human decision making in terms of speed and accuracy. The value creation potential is one of the greatest across industries, with AI potentially unlocking $1 trillion in annual value for banks.
In dozens of use cases, AI technologies can help drive revenue through greater personalization of services for customers (and employees); reduce costs through efficiencies from greater automation, lower error rates, and better use of resources; and uncover new and previously unrealized opportunities based on an improved ability to process and generate insights from vast amounts of data.
More broadly, breakthrough AI technologies can dramatically improve banks' ability to achieve four key outcomes: higher profits, personalization at scale, distinctive omnichannel experiences, and rapid innovation cycles. Banks that fail to put AI at the center of their core strategy and operations - what we call "AI-first" - risk being overtaken by competitors and abandoned by their customers. This risk is exacerbated by four current trends:
Rising customer expectations
Due to increasing adoption of digital banking. In the early months of the COVID 19 pandemic, use of online and mobile banking channels is estimated to have increased by 20 to 50 percent across countries, and is expected to continue at this higher level once the pandemic subsides. In various global markets, between 15 and 45 percent of consumers expect to make fewer branch visits once the crisis is over. As the use of digital banking services increases, so do consumers' expectations, especially when compared to the standards they are accustomed to from leading consumer Internet companies. Meanwhile, these leading digital experience providers are raising the bar on personalization, to the point where they sometimes anticipate customer needs before the customer perceives them and offer highly tailored services at the right time through the right channel.
The use of advanced AI technologies
By leading financial institutions is steadily increasing. Nearly 60 percent of financial services firms surveyed in the McKinsey Global AI Survey5 say their organizations have integrated at least one AI capability. The most commonly deployed AI technologies are: Robotic process automation (36 percent) for structured operational tasks, virtual assistants or conversational interfaces (32 percent) for customer service departments, and machine learning (25 percent) to detect fraud and support underwriting and risk management. While at many financial services firms the use of AI is selective and focused on specific use cases, a growing number of banking executives are taking a comprehensive approach to the use of advanced AI and deploying it across the lifecycle, from front to back office
Are displacing traditional financial services. By enabling access to a variety of services through a common access point, digital ecosystems have changed the way consumers discover, evaluate, and purchase goods and services. For example, WeChat users in China can use the same app not only to exchange messages, but also to book a cab, order food, schedule a massage, play games, send money to a contact, and access a personal line of credit. Similarly, non-bank companies and "super apps" in all countries are embedding financial services and products into their journeys, providing compelling experiences for customers and disrupting traditional methods of discovering banking products and services. As a result, banks need to rethink their participation in digital ecosystems and leverage AI to unlock the full power of data from these new sources.
Are pushing into the financial services sector as the next complement to their core business models. Globally, leading tech giants have built exceptional market advantages: a large and dedicated customer network, reams of data that enable a robust and increasingly accurate understanding of individual customers, natural strengths in developing and scaling innovative technologies (including AI), and access to low-cost capital. In the past, tech giants have aggressively moved into adjacent business areas to create new revenue streams and retain customers with new offerings. Big tech companies have already gained a foothold in the financial services sector in select areas (particularly payments and, in some cases, lending and insurance) and may soon seek to leverage their advantages to deepen their presence and build greater scale.
What could the AI bank
of the future look like?
To meet rising customer expectations and compete in the digital age with AI, the AI-first bank will provide offerings and experiences that are intelligent (i.e., recommending actions, anticipating and automating important decisions or tasks), personalized (i.e. relevant and timely, and based on a detailed understanding of past customer behavior and context), and truly cross-channel (seamless across physical and online contexts and across multiple devices, and providing a consistent experience), connecting banking functions with relevant products and services beyond banking. Figure 1 illustrates how such a bank might engage a retail customer throughout the day. Figure 2 shows an example of the banking experience of a small business owner or the treasurer of a mid-sized company.
Internally, the AI-first institution will be optimized for operational efficiency by making manual tasks extremely automated (a "zero-ops" mentality) and replacing or augmenting human decision-making with advanced diagnostic engines in various areas of banking operations. These gains in operational performance will result from the broad application of traditional and leading AI technologies such as machine learning and facial recognition to analyze large and complex reserves of customer data in (near) real time.
The AI-first bank of the future will also enjoy the speed and agility that characterize digital-native companies today. It will innovate quickly and roll out new features in days or weeks rather than months. It will work intensively with partners to deliver new value propositions that are seamlessly integrated with journeys, technology platforms, and data sets.
What are the barriers keeping banks
from deploying AI capabilities at scale?
Established banks are confronted with two objectives that seem contradictory at first glance. On the one hand, banks need to achieve the speed, agility and flexibility inherent in a fintech. On the other hand, they must continue to manage the scale, security standards and regulatory requirements of a traditional financial services company.
Banks' core technological systems, designed for stability, have proven their worth, especially in supporting traditional payments and lending. However, banks need to address several weaknesses in legacy systems before they can deploy AI technologies on a large scale. First and foremost, these systems often lack the capacity and flexibility to support the variable computational demands, data processing needs, and real-time analytics that closed-loop AI applications require.
Core systems are also difficult to change and require significant resources to maintain. In addition, many banks' data assets are fragmented across multiple silos (separate business and technology teams), and analytics efforts are narrowly focused on individual use cases. Without a centralized data backbone, it is virtually impossible to analyze relevant data and generate an intelligent recommendation or offer at the right time. If data is the bank's fundamental raw material, it must be managed and securely delivered in a way that enables the analysis of data from internal and external sources at scale for millions of customers in (near) real time at the "point of decision" across the enterprise.
Finally, to scale various analytics and advanced AI models, organizations need a robust set of tools and standardized processes to build, test, deploy, and monitor models in a repeatable and "industrial" manner. Investment in core technologies is critical to meet increasing demands for scalability, flexibility, and speed.
Banks' traditional operating models further hinder their efforts to meet the need for continuous innovation. Most traditional banks are organized around distinct business units, with centralized technology and analytics teams structured as cost centers. Business owners define their goals unilaterally, and alignment with the company's technology and analytics strategy (if any) is often weak or insufficient.
Siloed teams and "waterfall" implementation processes inevitably lead to delays, cost overruns and suboptimal performance. Companies also lack a test-and-learn mentality and robust feedback loops that encourage rapid experimentation and iterative improvements. Often dissatisfied with the performance of past projects and experiments, abandoned and experimenting dissatisfied, executives tend to get hung up on critical functions, neglecting skills and talents that should ideally be developed internally to differentiate from the competition.
How can banks transform
to implement an AI first approach?
To overcome the challenges that limit the enterprise-wide adoption of AI technologies, banks must take a holistic approach. To become AI-first, banks must invest in transforming capabilities at all four levels of the integrated capability stack: the engagement layer, the AI-powered decision layer, the core technology and data layer, and the operating model.
When these interdependent layers work together, they enable a bank to deliver distinctive omnichannel experiences to its customers, support personalization at scale, and drive the rapid innovation cycles that are critical for competitiveness in today's world. Each layer has a unique role to play - underinvestment in any one layer creates a vulnerability that can cripple the entire organization. The following sections explore some of the changes banks need to make in each layer of this competency stack.
Layer 1 redesign of the customer loyalty layer
Customers increasingly expect their bank to be present in their end-use journey, to know their context and needs wherever they interact with the bank, and to enable a frictionless experience. Many banking activities (e.g., payments, certain types of loans) become invisible because the Journeys often begin and end at interfaces outside the bank's own platforms. In order for the bank to be ubiquitous in customers' lives and solve latent and emerging needs to satisfy needs while providing intuitive omnichannel experiences, banks need to reimagine how they engage with their customers and make several key changes.
First, banks need to move beyond highly standardized products and create integrated offerings that target "things to do." This requires embedding personalization decisions (what to offer, when to offer, through which channel to offer) into the core customer journey and developing value propositions that go beyond the core banking product and incorporate intelligence that automates decisions and activities on behalf of the customer. In addition, banks should strive to integrate relevant non-banking products and services that, together with the core banking product, comprehensively address end-customer needs.
An example of the jobs-to-be-done approach is how fintech Tally helps customers manage multiple credit cards. The fintech's customers can solve multiple problems, including deciding which card to pay first (tailored to forecast monthly income and expenses), when to pay, and how much to pay (minimum balance or repayment) - a complex set of tasks that customers themselves often don't do well.
The second necessary shift is to seamlessly embed the customer journey into partner ecosystems and platforms, so that banks engage customers at the point of end-use, leveraging partner data and channel platforms to drive engagement and usage. ICICI Bank in India embedded basic banking services into WhatsApp (a popular messaging platform in India) and reached one million users within three months of launch. In a world where consumers and businesses increasingly rely on digital ecosystems, banks should
decide what stance they want to take across multiple ecosystems - i.e., develop themselves, orchestrate, or partner - and adjust the capabilities of their engagement layers accordingly.
Third, banks need to redesign the entire customer experience and the specific journeys for omnichannel interactions. This includes enabling customers to move seamlessly across multiple channels (e.g., web, mobile app, branch, call center, smart devices) within a single journey and maintaining and continuously updating the current context of the interaction. Leading consumer Internet companies with offline-to-online business models have reshaped customer expectations in this dimension. Some banks are driving omnichannel journey design, but most will need to catch up.
Redesigning the AI bank engagement layer requires a clear strategy for how to engage customers through non-bank partner channels. Banks need to be flexible in developing experiences inside, outside, and beyond the banking platform by designing interaction interfaces to enable and personalization for customers, reengineering backend processes, and ensuring that data capture funnels (e.g., clickstream) are
are granularly embedded in the bank's interaction layer. All of this is aimed at enabling granular understanding of Journeys and continuous improvement.
Layer 2 Building the AI-supported decision-making layer
To deliver personalized messaging and decisions to millions of users and thousands of employees in (near) real time across the full spectrum of interaction channels, the bank needs to develop an AI-powered decision layer at scale. Across the bank, AI techniques can either fully replace or complement human judgment to deliver significantly better outcomes (e.g., greater accuracy and speed), improve the customer experience (e.g., more personalized interactions and offers), provide actionable insights to employees (e.g., which customer to contact first with recommendations for the next best action), and strengthen risk management (e.g., earlier detection of default probabilities and fraudulent activity).
To establish a robust AI-powered decision layer, banks must move from developing specific use cases and point solutions to an enterprise-wide roadmap for deploying advanced analytics (AA)/machine learning (ML) models across entire business units. To illustrate, more than 20 lifecycle decisions can be automated in the unsecured consumer loan space alone. To enable decision model development at scale, banks must make the development process repeatable, enabling them to deliver solutions effectively and on time. This requires close collaboration between business teams and analysts, as well as robust model development tools, efficient processes (e.g., for code reuse across projects), and knowledge dissemination (e.g., repositories) across teams. In addition to developing decision models at scale across different domains, the roadmap should also include plans for embedding AI into the normal business process. This effort is often underestimated and requires a rewiring of the business processes into which these AA/AI models will be embedded, an "explanation" of AI decision making to end users, and a change management plan that addresses changes in employee mindset and missing capabilities. To foster continuous improvement beyond initial deployment, banks also need to have infrastructure (e.g., data measurement) and processes (e.g., periodic performance reviews, risk management of AI models) in place to allow feedback loops to flourish.
In addition, banks need to extend their homegrown AI models with rapidly growing capabilities (e.g., natural language processing, computer vision techniques, AI agents and bots, augmented or virtual reality) in their core business processes. Many of these cutting-edge capabilities have the potential to create a paradigm shift in customer experience and/or operational efficiency. While many banks may lack both the talent and the investment readiness required to develop these technologies themselves, they must at least be able to rapidly source and integrate these emerging capabilities from specialized vendors through an architecture enabled by an application programming interface (API), encourage continuous experimentation with these technologies in sandbox environments to test and refine applications and assess potential risks, and then decide which technologies to deploy at scale.
To deliver these decisions and capabilities and address customers across the lifecycle, from acquisition to upsell and cross-sell to retention and win-back, banks need to build an enterprise-wide digital marketing machinery. This machinery is critical for translating decisions and insights generated at the decision level into a series of coordinated interventions delivered through the engagement level of the bank. This machinery has several critical elements that include:
Data processing pipelines that capture a range of data from multiple sources both within the bank (e.g., clickstream data from apps) and beyond (e.g., third-party partnerships with telecom providers)
Data platforms that aggregate, develop and maintain a 360-degree view of customers and enable near real-time execution of AA/ML models
Campaign platforms that track past actions and coordinate predictive interventions across interaction layer channels
Layer 3 Strengthening core technology and data infrastructure
Deploying AI capabilities across the enterprise requires a scalable, resilient, and adaptable set of core technology components. A weak core technology backbone that lacks the investment required to modernize can dramatically reduce the effectiveness of the decision and engagement layers. The core technology and data layer consists of six key elements:
1. Tech forward strategy banks should have a unified technology strategy that is closely aligned with the business strategy and makes strategic decisions about which elements, capabilities and talent the bank will keep in-house and which it will source through partnerships or vendor relationships. In addition, the technology strategy must outline how each component of the target architecture both supports the bank's vision of being an AI-first institution and interacts with each layer of the capability stack.
2. Data management for AI supporting world The bank's data management must ensure data liquidity, i.e. the ability to access, ingest and manipulate the data that serves as the basis for all insights and decisions generated in the decision layer. Data liquidity increases as functional silos are eliminated, allowing multiple departments to work with the same data and coordinate more effectively. The data value chain begins with seamless sourcing of data from all relevant internal systems and external platforms. This includes ingesting data into a data pool, cleansing and tagging data needed for various use cases (e.g., regulatory reporting, scale-level business intelligence, AA/ML diagnostics), separating incoming data (from both existing and potential customers) that should be made available for immediate analysis from data that needs to be cleansed and tagged for future analysis. In addition, as banks design and build their centralized data management infrastructure, they should develop additional control and monitoring tools to ensure data security, privacy, and regulatory compliance - e.g., timely and role-appropriate access within the organization for various use cases.
3. Modern API architecture APIs are the link that enables controlled access to services, products, and data, both within the bank and beyond. Within the bank, APIs reduce the need for silos, increase the reusability of technology assets, and promote technology architecture flexibility. Outside the bank, APIs accelerate the ability to partner externally, unlock new business opportunities, and enhance the customer experience. While APIs can unlock significant value, it is critical to first define where they will be deployed and establish central governance to support their development and deployment.
4. Intelligent infrastructure As enterprises across industries increase the share of workloads handled by public and private cloud infrastructures, there is ample evidence that cloud-based platforms enable the higher scalability and resilience that are critical to an AI-first strategy. In addition, cloud-based infrastructure lowers IT maintenance costs and enables self-service models for development teams that enable rapid innovation cycles (e.g., setting up new environments in minutes rather than days) through the provision of managed services.
Layer 4 Switching to a platform operating model
The AI-first bank of the future requires a new operating model for the organization to achieve the agility and speed required to unlock value at all levels. While most banks are transforming their technology platforms and assets to become more modular and agile, work teams within the bank continue to operate in functional silos under suboptimal collaboration models and often lack alignment of goals and priorities.
The platform operating model envisions cross-functional business and technology teams organized as a series of platforms within the bank. Each platform team controls its own assets (e.g., technology solutions, data, infrastructure), budgets, performance metrics, and talent. In turn, the team delivers a family of products or services either to end customers of the bank or to other platforms within the bank. In the target state, the bank could end up with three archetypes of platform teams. Business platforms are customer- or partner-focused teams that focus on achieving business outcomes in areas such as consumer lending, corporate lending, and transactional banking. Enterprise platforms deliver specialized capabilities and/or shared services to establish standardization across the organization in areas such as collections, payment processing, human resources, and finance.
Platforms enable enterprise and business platforms to deliver cross-cutting technical functionality such as cybersecurity and cloud architecture.
By integrating business and technology into common platforms operated by cross-functional teams, banks can break down organizational silos, increase agility and speed, and improve alignment of goals and priorities across the enterprise. Becoming an AI-focused bank requires transforming capabilities at all four levels of the capability stack. Ignoring challenges or underinvesting in one level impacts all levels and results in a suboptimal stack that is unable to achieve business goals.
A practical way to start is to evaluate how the bank's strategic goals (e.g., growth, profitability, customer retention, innovation) can be significantly supported by the various AI technologies - and to dovetail the AI goals with the bank's strategic goals. Once this alignment has occurred, bank executives should conduct a comprehensive diagnostic of the bank's starting position at all four levels to identify areas that need key changes, additional investments, and new talent.
They can then translate these insights into a transformation roadmap that includes business, technology, and analytics teams. Equally important is developing an execution approach that is tailored to the organization. To ensure sustainability of change, we recommend a two-pronged approach that balances short-term projects that deliver business value on a quarterly basis with iterative building of long-term institutional capabilities. In addition, depending on their market position, size, and ambitions, banks may not need to build all capabilities themselves. They could choose to keep core differentiating capabilities in-house and acquire non-differentiating capabilities from technology vendors and partners, including AI specialists.
For many banks, adopting AI technologies across the enterprise is no longer an option, but a strategic imperative. Critical to success is looking at and building the bank's capabilities holistically across the four tiers.