This blog is part of Accelerating Frontier Transformation, a four-part series exploring how organizations are turning AI into business value. Based on conversations with industry leaders at Microsoft Frontier Tours, the series looks at AI’s impact across employee experience, customer engagement, business processes, and innovation. This third post focuses on reshaping business processes.
Rethinking how work gets done
AI can help us prepare a document, analyze information, or summarize a meeting. But its potential extends to reshaping entire business processes. To realize more value from AI, organizations are rethinking how work is organized across teams and functions: how knowledge informs decisions, how people and agents collaborate, and where human expertise and accountability are needed. This creates an opportunity to improve business outcomes through new ways of operating, rather than simply completing the same tasks faster.
The leaders we spoke with at several Microsoft Frontier Tours around the world described how this shift is taking shape. Their examples ranged from bringing organizational knowledge into customer proposals to using AI-supported forecasts in business decisions. They also shared ambitions to move beyond personal productivity and change how their organizations deliver value.
Reshaping business processes is a key part of Frontier Transformation, built on intelligence and trust. Intelligence can ground AI in the knowledge and context that work requires. Trust means building security, governance, and human oversight into how AI is designed and used.
Starting with the outcome
A faster process is useful, but speed alone does not define its value. What matters is what the organization needs that process to deliver: a stronger customer proposal, a better-informed decision, or more capacity for work that requires professional expertise. That outcome gives teams a reason to reconsider how the work is organized, rather than simply add AI to the steps already in place.
At Advania UK, an IT services provider, answering a customer request for proposal (RFP) means showing how the company’s capabilities meet that customer’s requirements. The organization is using AI to change how its team prepares these responses, focusing on quality and turnaround time as priorities.
Our RFP agent is helping the bids team to become incredibly more efficient. But it’s not just the speed which we were after, is the quality as well, in which we can review many thousands of previous RFPs and quickly assemble very well structured and comprehensive responses to RFPs as they come in now, in less than half the time that we used to.
Nick Isherwood, Chief Operating Officer, Advania UK
For Trustmarque Ultima, a technology provider that helps organizations with adoption and change management, identifying where to apply AI begins with understanding the customer’s business needs. Its work with customers shows the importance of defining relevant use cases before choosing an application:
A lot of the customers that I speak to, they’re still at the persona mapping and use case mapping stage to actually understand where it would benefit them best, and this is why change management and actually speaking to end users is so important.
Fajar Raja, Customer Adoption Director, Trustmarque Ultima
Those conversations can help connect a business priority to the work that needs to change. The starting point is a defined need within the business, rather than adopting AI without a clear purpose.
Connecting knowledge and building trust
Once the outcome is clear, AI needs access to the information that makes the work meaningful. A process may depend on organizational experience, operational data, or knowledge held across several teams. Connecting that context can help people and agents contribute to the same business objective.
At Stanwell Corporation, an energy generator in Queensland, the amount of data underpinning renewable energy has made AI important to energy trading. Putting AI into this work also requires technology that the business can maintain and use beyond a single application:
We need to build a standardized platform that is well supported, well maintained, and that is used across the enterprise, such that we don’t need to support different technology stacks.
Kevin Lin, Chief Information Officer, Stanwell Corporation
Stanwell built that foundation on Microsoft Azure, supported by its technology team. One application is energy-price forecasting, which Stanwell uses to help inform trading decisions. The usefulness of those forecasts extends to other parts of the business:
It also helps from managing our portfolio risk perspective, and it also helps from a financial forecast perspective.
Kevin Lin, Chief Information Officer, Stanwell Corporation
Together, these examples connect the technology foundation to the work it enables. At Stanwell, AI-supported analysis is becoming part of processes that support trading, risk assessment, and financial forecasting, rather than an output separate from its business processes.
Connecting organizational knowledge also requires confidence in how it will be handled. GHD, a global professional services firm working across engineering, architecture, and environmental science, made that consideration part of its approach to putting AI to work. Its leaders described the clarity they needed before employees could use AI with business information:
The clarity that we had around enterprise data protection, the degree to which access control limitations and privacy limitations follow the user and the ability for the user’s responses to then be protected in that same way was critical.
Paul Murphy, Chief Information Officer, GHD
GHD also helped employees understand sources, references, and where errors might occur in AI responses. That addresses another part of trust: people need to know how to assess the information they use, not simply have access to it. Data protection, access controls, and informed human review all matter when AI-supported work contributes to a customer response or business decision.
Proving value before expanding
The value of a redesigned process should be visible in the work it supports. Turnaround time matters, but so do quality, cost, and the capacity people have to apply their expertise. Looking at these measures together helps leaders understand whether a change is delivering the outcome they intended.
For Grant Thornton Australia, an accounting, advisory, and consulting firm, attracting and retaining talent was an important part of the case for AI. The organization introduced its nine-day working schedule over two weeks in March 2023 and made it permanent after a year-long trial. The model combines 100% pay, 90% of working hours, and 100% of expected outcomes. Changing how employees organize their work is therefore tied to maintaining what the firm delivers to clients. Within this broader working model, Grant Thornton has also explored how AI can support employees in managing their time while maintaining delivery expectations.
Those productivity gains, those time savings: AI was a real enabler of empowering our people to do that.
Jemma Hirst, Head of Digital Capability & Enablement, Grant Thornton Australia
The organization also examined the relationship between Copilot licensing and employees’ participation in recharge time, or time off, under that working model:
People who had a Copilot license were taking 10% more recharge time than those who did not have the license.
Jemma Hirst, Head of Digital Capability & Enablement, Grant Thornton Australia
This customer-reported association offers one way to examine AI use within a changed working model, alongside delivery expectations, but does not establish that Copilot caused the difference in recharge time. Other organizations will need measures suited to their own processes, such as response quality and preparation time for customer proposals, or how forecasts inform business decisions. Establishing those measures from the outset gives leaders a basis for improving the workflow and deciding what is ready to expand.
Across these conversations, reshaping business processes begins with a clear purpose, develops through relevant knowledge and trust, and progresses through evidence from real work. The aim is not simply to deploy AI more broadly, but to build new ways of operating that deliver business value.
Make three practical moves now
- Start with the outcome, then rethink the process. Define the business result you want to improve. Work with the people involved to identify how AI could change the workflow, not just speed up individual tasks.
- Connect knowledge and build in trust. Bring relevant information and expertise into the workflow, with clear access permissions, accountability, and human review where decisions require it.
- Prove value before expanding. Measure changes in speed, quality, cost, and capacity against the outcome you set. Use what you learn to improve the process and determine what can be reused across the business.
Learn more
- Visit the Frontier Transformation site to explore how you can put AI to work, build trust, and scale transformation with Microsoft across everyday workflows.
- Download the e-book Four Paths to Business Value with AI to learn how you can move from AI experimentation to Frontier Transformation with intelligence and trust.
- Read the AI Strategy Roadmap for guidance on aligning AI with business goals and building the foundations to scale.
Next in the Accelerating Frontier Transformation series: Bend the curve on innovation.
