At Davos this year, the hallways were abuzz with talk of agentic AI: small intelligent systems that take actions and automate tasks. The pace of innovation in the world of AI can be overwhelming for business leaders and policy-makers, and many are unsure how to act on the opportunity. While these concepts may seem complex, business leaders can use a simple framework to navigate and unlock the value of agents.

Much of the conversation so far has focused on agents driving productivity, but the real impact goes beyond that. Agents will make work easier, faster and more effective – whether by handling tedious manual tasks, speeding up handovers between teams, or accelerating market delivery. These improvements drive speed to value – this is how we should measure the outcome of our investment in agentic systems. It’s not about efficiency for its own sake; it’s about how quickly value is delivered.

To get the most from agents, they should be integrated across the entire value chain. Here’s a straightforward framework to help you cut through the complexity and unlock value for your business.

The Discover, Decide, Deliver framework is a useful model to frame how business leaders should think about their investment and approach to adopting AI and agentic workflows.

Let’s explore the process of bringing a revolutionary medicine to market with agents using Discover, Decide, Deliver. Traditional drug development is a long and costly journey, often taking several years. Agentic AI can automate repetitive manual steps and create more seamless transitions between stages, accelerating time to market.

The process begins with Discovery. This is where initial research takes place to understand the disease, decide which aspect of disease biology to target, and then develop a compound that will alter that aspect of the disease. It also includes data analysis to test molecules and compounds. Next is the Decision phase, where researchers and clinicians land on how and where the drug will be used, from speeding up new clinical trial protocols to analyzing massive amounts of complex data from these trials. Then there’s the Delivery phase, which involves producing the drug, educating patients and doctors, and bringing it to market.

While discovery is happening faster than ever, manual processes between and within stages can cause delays, errors and miscommunications, slowing down overall progress. In this case, speed to value isn’t just a metric – it’s the difference between getting a drug to market faster or losing market share. For instance, one of the world’s largest pharmaceutical manufacturers, Novartis, applied Generative AI in its drug development process even before agents came along. Imagine what they could do now. Ideally, agents can cut timelines from decades to years and years to months, enabling faster decisions, delivery and learning. Here, agents don’t just help you do more with less – they accelerate value by getting you to the right outcomes faster.

This concept applies across industries. For example, Levi Strauss & Co built a powerful but complex traditional AI system to detect fashion trends. Agents could optimize this process by analyzing online images daily, tracking the shift from fitted to baggy jeans across demographics and delivering insights faster. The Coca-Cola Company is taking it a step further by using AI to enhance its relationships with retailers. Imagine an agent that gives recommendations on how to arrange products in coolers based on current consumer preferences. In these cases, speed to value is achieved by 1) accelerating the transition from analysis to action and 2) reducing the delay between insights and action across companies.

In a recent article, we explored the evolving role of leadership in the age of AI. The rise of agents requires a similar mindset shift. Below, we apply the Discover, Decide, Deliver framework to help leaders ask the right questions in an agent-driven world.

The key to leveraging agents in discovery is understanding your objective and using the right data to fuel faster insights and analysis from more places. Agents can quickly pull data from multimodal sources – text, audio, visual – and spot patterns that reveal opportunities. Using reasoning capabilities and accelerated computing, they help connect the dots faster, turning data into answers.

Leaders need to ask:

Across industries, leading companies like Netflix have built their success on high-velocity decision-making. They thrive because they minimize the iteration cycle and execute quickly to deliver innovation. Agents that can reason allow any organization to streamline decision-making, reduce the cost of experimentation and take action faster. With agents handling manual, repetitive, error-prone tasks, leaders can focus on more strategic decisions sooner, accelerating the innovation cycle.

Leaders need to ask:

Today, the speed of delivery depends on how well you collaborate with customers, vendors and partners. Agents enable seamless collaboration across organizations, speeding up delivery by ensuring that processes are in sync between teams, suppliers and customers (just like APIs did for digital transformation).

Leaders need to ask:

Agentic AI is changing how businesses drive outcomes by accelerating speed to value. The companies that will dominate the next decade won’t just use AI – they’ll fully integrate agent-driven actions across their business and partnerships. To capture the benefits of speed to value, this will need to be both a process change and a mindset change across leadership and throughout the organization.

How to get started:

Success comes down to implementing a scalable and repeatable framework that works within your existing processes. Discover, Decide, Deliver is one example that will help you break through the complexity and unlock the promise of agents.

https://www.weforum.org/stories/2025/03/ai-agent-business-value/