The bi-modal imperative

AI has turned up the pressure on the C-suite. CEOs field relentless questions about AI strategy, competitors announce sweeping AI transformations and investors inundate earnings calls with questions about AI initiatives.

Yet for all the fanfare and investment, companies are discovering an uncomfortable truth. Despite implementing AI tools, deploying chatbots and appointing Chief AI Officers, the transformational impact they were promised simply isn’t materializing. According to McKinsey, while 78% of organizations report using AI, over 80% of them saw no tangible enterprise-level impact on earnings. A study by BCG paints a similar picture: only 4% of companies create substantial value from AI initiatives, while 74% struggle to show any value at all.

According to the 2025 AI Business Predictions by PwC, 49% of technology leaders already claimed that AI was “fully integrated” into their companies’ core business strategy. That is an impressive but misleading number. While companies are undoubtedly integrating AI, they are largely doing it to do more of the same work, only now with AI.

We’re seeing the same mistake businesses made during the early days of digital transformation: using new technology to do old things slightly better rather than reimagining what’s possible. Compounding the issue, in the race to build the best large language models, organizations focus on data quantity and quality, overshadowing critical gaps in governance and problematic data.

But it’s not too late to get it right by harnessing AI to capitalize on its transformative potential. There’s a clear path forward, and it starts with understanding that AI transformation isn’t a single strategy but a dual one, centered around optimization and innovation.

Banking on two modes

I’ve been struck when speaking to customers in the finance sector how change doesn’t happen overnight and isn’t all or nothing. Financial institutions have seen breakneck-speed innovations and sweeping operational changes. Many customers have welcomed the change, carrying out transactions online and never going into a branch or interacting directly with an employee. Yet others continue to do everything in person, expect physical letters signed by the branch manager and want a teller to guide them. As a result, financial institutions have found themselves improving their current operations while also introducing technological innovations simultaneously.

This extends beyond consumer behavior into operational technology strategy, exemplified by the concept of bimodal IT that Gartner introduced back in 2014. The core tenet of a bimodal strategy is to split focus between two distinct modes. Mode 1 focuses on stability, predictability and maintaining core systems. Mode 2 emphasizes agility, speed and innovation, often involving new digital initiatives and customer-facing applications.

As the financial industry has shown us, Modes 1 and 2 are not mutually exclusive. This is the essence of bimodal thinking: rather than thinking of the modes as two consequent steps, businesses should invest resources, strategy and measures in parallel, pursuing bimodal outcomes in performance and productivity optimization and value creation. With AI, this framework is more relevant and useful than ever.

Changing the wings while flying

In the rush for AI adoption, innovation and transformation, companies will need to run their existing business while leveraging AI to free up resources for organizational transformation, which will then drive business transformation. For many companies, the next two years will be the ultimate “change the wings while the plane is flying” exercise.

When evaluating AI objectives, the first step is to identify the prime use cases that extend beyond single-modal approaches. These strategies can be defined as:

• Reduce bottom-line costs. Implement agentic solutions to automate and increase efficiency, remove friction and optimize throughput.
• Increase top-line revenue. Use AI to maintain production lines, optimize supply chains and streamline processes.
• Improve satisfaction. Enhance experiences for employees, customers and the entire ecosystem to improve satisfaction and the ease of doing business.
• Drive business transformation. Create capacity for entirely new business units, service offerings, products or geographical expansion.

These are not mutually exclusive. Organizations can pursue single or multiple outcomes through a bimodal approach that enables parallel strategies. As Dan Priest, PwC US chief AI officer, notes: “Top performing companies will move from chasing AI use cases, to using AI to fulfill business strategy.” Like aircraft engineers redesigning and swapping out wings mid-flight, organizations must maintain operational altitude while simultaneously building capabilities that will define their competitive trajectory.

Mode 1: Securing your mission-critical operations

To evaluate your strategy, you must first establish what is essential to the business and use AI to operationalize effectiveness in Mode 1. Mode 1 focuses on establishing and optimizing your business-critical operations. These are the things that provide the core value to create the foundation for innovation. What simply cannot fail? What is generating the core revenue? This could be your production line, telecommunication network, sporting events, citizen services system or anything that drives your revenue and purpose.

In Mode 1, you want to apply AI to process improvement and optimization. What delivery bottlenecks could AI eliminate? Which customer or employee frustrations could be resolved? Could AI remove any of these constraints? The application of AI thinking to Mode 1 operations will reduce costs, increase revenue and improve operational experiences.

Mode 2: Boundless innovation

While Mode 1 creates operational excellence, Mode 2 transforms efficiency gains into exponential growth opportunities. This is where technology makes you faster, more agile and more innovative. The personnel, capital and capacity freed through Mode 1 optimization become the fuel for Mode 2 innovation.

This parallel approach enables new project launches, geographical expansion, product development and service line creation. Your strategic vision should drive Mode 2 initiatives, which will evolve alongside AI technologies and resulting capabilities. Mode 1 achieves linear gains through performance improvements and cost savings. Mode 2 unlocks exponential output.

It is important to note that this exponential output is possible without exponential growth. An AI-first company prioritizes bringing new offerings to the market and dramatically growing revenue without necessarily increasing its size. A recent McKinsey report on micro-, small- and medium-sized enterprises (MSMEs) highlights that some MSMEs from 2000 “now represent 17 percent of publicly traded companies valued at $10 billion or more as of 2023.” Tech companies show particularly strong performance — nearly 25% of large public tech firms were MSMEs within the past quarter-century. The lesson here is that nimble organizations accelerating innovation cycles will amplify transformation opportunities.

The intelligence revolution, by the people and for the people

The two strategies cannot thrive in isolation. What you wish to achieve in the future, potentially by using a Mode 2 strategy, should influence the design and implementation of your Mode 1 strategy. Have a plan for what those efficiency gains will be used for and prioritize building an organizational foundation that fosters innovation so you can put Mode 2 into action as you carry out Mode 1.

In “How AI can drive business transformation,” I argued that companies need to strategically reallocate the time freed up by AI rather than simply seeing it as an opportunity for cost-cutting. AI-generated efficiencies have the potential to reinvent your organization through upskilling, skills mapping and innovation initiatives that prepare businesses for an AI-driven future. Now is the time to foster environments where employees feel empowered to explore AI applications, take calculated risks and learn from both successes and failures.

We are in the midst of an intelligence revolution that offers unprecedented opportunities to embrace both optimization and innovation simultaneously. Your AI strategy isn’t about AI—it’s about how AI helps you achieve exponential performance, relevance and market leadership. Organizations mastering this dual approach will define the next generation of business.

This article is published as part of the Foundry Expert Contributor Network.
Want to join?