Strategize: Mapping and accelerating innovation
The first phase of our generative AI adoption framework involves establishing a “North Star”—the core strategic objectives that will drive activities forward—as well as defining guardrails that can accelerate innovation and creating a map of evolving projects.
Together, these steps will ensure effective innovation and prioritization of projects that herald the most promise for the future.
- Establish a “North Star”: The allure of new technologies like generative AI can lead to a rush toward innovation without a clear path or purpose. Business leaders need to recognize this challenge and work diligently to align generative AI investments with their business's overarching corporate strategy.
In some cases, generative AI adoption might not only enhance the existing business model but also transform it entirely, leading to new ways of operating, creating value and engaging with customers. Ultimately, generative AI adoption may act as the fulcrum for managing cultural change throughout the organization as the use of the technology becomes more pervasive.
By creating a well-defined strategy, businesses can define a guiding North Star to which the constellation of other initiatives and activities across the business can connect. This serves as a reference point, ensuring that all efforts are coordinated and aligned.
- Pursue risk-conscious innovation: Every business has its own unique risk parameters—and all explorations with a nascent technology like generative AI should occur within the appropriate boundaries. The goal isn't to hamper innovation at the fringes of the business; instead, it's about building a governance and ethical framework that encourages effective innovation.
In the earliest stages of generative AI adoption, these guardrails will be informed by regulatory and compliance commitments, information security protocols and corporate processes.
For example, healthcare organizations will need to be especially protective of—or even prohibit—the use of patient data, while finance organizations might be particularly way of involving core systems lest they fall foul of always-on requirements. Such restrictions could change as businesses’ understanding of generative AI evolves and the impact on core systems and sensitive data is better known.
Whatever the guardrails, the aim is to clearly communicate them to stakeholders across the business. But crucially, it's a delicate balancing act—freeing individuals and teams to move forward with experimentation as long as they stay within the defined parameters.
- Find connections across use cases: It's essential to collect business cases from various parts of the organization and chart out a clear constellation map of these projects as they evolve.
The reason the constellation map will become particularly crucial is that as use cases evolve, the real business value will emerge from connecting disparate initiatives that add value to each other.
Take for example, the hypothetical case of an automotive sales team building a generative AI-driven platform to develop customer insight reports from vast volumes of structured and unstructured data to tailor aftersales solutions that meet evolving customer needs.
Meanwhile, product development teams are working on their own gen AI system to enhance new vehicle designs. Separately, both programs add value to their respective teams. However, by grouping the two engagements together, the product development effort could be fueled with real-time customer feedback to enhance designs, and the sales team could be armed with insights into new products as they evolve. The value is multiplied.
Identifying these connections may not happen immediately. But by creating use-case constellations, businesses can unlock transformative potential across business functions, new ways of working and even new business spin-offs.
For example, in the finance sector, generative AI use cases might encompass areas such as fraud detection, risk management and customer service. By linking these use cases, a financial institution could create a specialized business unit—and even a completely new industry—dedicated to AI-driven financial security and regulatory compliance and oversight.
Prove: Turning on the value amplifier
Once the business establishes a defined North Star, a clear constellation map of projects across the business, and clear yet evolving guardrails to structure innovation, the focus then turns to building a value amplifier.
In practice, this takes the form of a cross-functional team fielding experts from a wide range of functions, including HR, learning and development, IT, compliance and corporate strategy.
This team should guide business leaders as they organize pilots, design experiments and develop proofs of concept. But more importantly, this group should help inform and direct the overarching strategy, prioritization, success measurement and coordination efforts. It should also help to finetune business cases by applying learnings and results from mature projects to those in their infancy.
Crucially, given the highly transformative nature of the technology, the team should also help shape the company-wide transformation, from the development of new revenue streams to the wholesale evolution of the operating model. Provided with an early glimpse into experiments as they shape up, the team can more closely examine the broader implications as they scale and help to manage change and cultural transformation as the organization evolves.
For example, assessing the variety of generative AI projects in action, the team might discover that certain skillsets are falling out of vogue faster than others; for example, system administration needs might shift into prompt engineering. Using this information, the team can begin structuring reskilling efforts to mitigate the impact of displaced workers, while developing scarce skillsets.
Scale: Transforming experiments into market-ready realities
To deliver value, experiments must scale. This will require resources (from compute power to monetary investment), a range of highly-skilled people and the capacity to evolve propositions as challenges emerge.
Throughout the scaling stage, the goal is to mesh the variety of projects rolling out across the business into a well-resourced program of initiatives that are effectively prioritized, both in terms of their impact and their potential to interlink with other projects for multiplied value.
Going back to the automotive example, as the product design team accelerates its plans, they’ll need more live customer feedback created by the sales team. Without the latter, the real-time development of new features will be misaligned with the realities of the market. But with a centralized approach to understanding both projects, it’s possible to manage activities and redirect resources to ensure joint success.
Reaching this point calls for a wide range of new skills—some of which may still be in their infancy as the labor market evolves. Acquiring the right talent to execute on these complex initiatives will likely require a combination of a robust partner strategy that enables projects to move at pace and scale, best-in-breed toolsets and reskilling and upskilling initiatives.
By strategically enhancing their existing resources through targeted training and development, organizations can create a flexible and skilled team capable of adapting to the evolving demands of various projects. Collaborative efforts with partners, combined with internal talent development, ensure that the right expertise is available at the right time, aligned with the dynamic needs of the business.
As more projects move into the scaling phase, they should be built into an iterative program of work that effectively maps timelines, resource allocation and skills as workers move from one project to the next and gain new skills and expertise. For example, a team that successfully supported the development of the new automotive sales tool can turn their attention to a new pitch from the marketing team looking at personalizing their own materials.
Improve: Fine-tuning strategy with real-world experiences
Finally, businesses need to collate all the data, experiences, successes and failures of every use case that's passed through the framework. By doing so, they can inform and evolve the strategy to accelerate and amplify the value of future cases.
The reality is, what looks like success at the start of a generative AI project will likely be manifestly different from what success looks like at the end. For example, a project with the initial goal of reducing costs or boosting efficiency could, when scaling, turn out to cost more than the system it’s replacing; however, it could also show potential for opening up new ways of generating revenue or enhancing customer experiences. Insights generated from past experiences could inform future use-case KPIs at the project’s outset.
The same is true for how projects are resourced as they move through the scaling phase; applying these learnings could lead to a more intelligent allocation of resources to deliver more value for less expense.
But, again, the greatest value is generated from a crystal-clear view of an evolving technology’s impact on business and operating models. By having a front-row seat to every project in every stage of development, leaders can examine how even the smallest business case can fundamentally alter how the company will operate in the future. They may be able to imagine an entirely new future state, new ways of working and new goals because their oversight of generative AI exposed and enabled a new realm of possibility.
The governance and ethics guardrails will likely also relax and contract based on new insights from the field. Data sets that were ringfenced in the early days of exploration may be liberated for certain use cases once security and compliance teams have a chance to examine the technology's impact, or even become more closely guarded.
The end-state is a framework that evolves and adapts as hundreds of initiatives move through all of these stages.
Repeat: Turning a framework into a value multiplier
Our generative AI framework is a holistic approach to managing hundreds, if not thousands, of use cases and proofs of concept. It helps the business map the flow of data, experiences and resources, and multiply the value of every project by combining constellations of initiatives into a comprehensive transformation roadmap. In short, it helps business leaders move beyond a point-solution-focused way of thinking, and truly embrace the potential of generative AI.
The truth is, most business leaders are still figuring out generative AI’s potential impact and ability to sustain real business value. It may ultimately prove to be an existential threat as nimble "AI natives" leapfrog slower moving incumbents. Or it may prove to be the salvation for data-rich intermediaries looking for effective paths to monetization.
What unites every business, however, is an urgent need to experiment with and explore what generative AI can bring to their business—and turn the hype and enthusiasm into a source of real business value.
For a deeper look at our insights into generative AI, see our report “Gen AI and the future of work: what businesses need to know.”
To learn more, visit the Consulting and Generative AI sections of our website, or contact us.