Launching AI Pilots: Why Change Management Will Keep You Ahead In The Race

Artificial intelligence (AI), and in particular Generative AI (GenAI), remains a hot topic in the boardroom and a high-priority investment area, however many organisations are highly risk averse and significantly budget constrained.

Taking a leap of faith toward new technology can seem like a risky move, but stagnating isn’t a safer option. As technology continues to evolve, it’s essential that businesses evolve too. AI is not just a buzzword; it has the potential to transform how organisations operate, driving game-changing productivity gains and revealing new avenues for topline growth.

With 72% of top-performing CEOs agreeing that competitive advantage will depend on who has the most advanced GenAI¹ and 71% of UK leaders believing GenAI will substantially transform their organisation within the next three years², a common question many Execs are asking is “how do we get started?”

Here we have outlined 5 tangible steps organisations can take to get their AI wheels in motion and past the start line whilst a more robust AI Strategy is developed. Some of these may seem obvious, but they are crucial in navigating a successful AI journey. In particular, strong change management capabilities which are often overlooked but essential for scaling solutions, realising material value and and staying ahead in the race.

1. Define Success

Success begins with a clear definition of strategic intent. A recent study found 81% of CEOs believe that inspiring their team with a common vision produces better outcomes¹. This highlights the importance of organisations ensuring that everyone is aligned on what AI development and investment success looks like. This clarity is also crucial to later support the establishment of foundational capabilities in people, technology, and data.

2. Use Case Selection

Identifying AI use cases that address priority business problems or support strategic opportunities is vital. These core value use cases should be business-led, scalable, and reusable across different operating companies. They also don’t need to be customer use cases; internal uses case can often reduce risk and accelerate progress.

By supporting the business in this selection process, organisations can ensure that AI initiatives are directly aligned with strategic goals and are positioned for maximum impact.

Case Study: Identifying an AI use case that addresses a priority business problem.

Challenge: Execs of a large retailer presented a priority problem statement to a target business area where there was energy and curiosity to innovate with AI. The problem was that customers couldn’t easily find products they were looking for.

Solution: The team created an AI-powered image recognition tool that analyses photographs shared by customers and returns either exact matches of the product or similar recommended products to buy online or in-store.

Result: Customers can easily find products they are looking for. Sales increased because of alternative product recommendations and complimentary item suggestions, whilst a seamless, personalised and interactive shopping experience improved customer engagement, satisfaction and loyalty.

3. Pragmatic Risk Management Balancing risk versus reward pragmatically is key. 71% of UK CEOs think trusted AI is impossible without effective AI governance, however only 35% feel they have good AI governance in place today¹. At the same time, 64% say they are willing to take more risk to maintain a competitive edge.

Creating a central AI governance team with the right mindset and culture is fundamental for managing risk effectively. This team should agree on ideal principles and behaviours that will guide them to understand, identify and mitigate risks while establishing standards that support responsible scaling. But it is equally important they are willing to take calculated risks, approve AI pilots and encourage fast learning. Maintaining momentum from ideation to mobilisation and productionisation is critical for success and organisations should avoid imposing excessive red tape, which can slow progress and stifle innovation.

Case Study: Defining principles and behaviours for effective AI governance.

Challenge: An AI governance team needed to evolve their mindset and develop a team culture that supported both effective AI risk management and value realisation.

Solution: The team defined key principles and behaviours they were committed to applying to their ways of working. These focused on being supportive, pragmatic, responsible, action-oriented and transparent. Each month the AI governance team scored themselves against each key principle to determine how well they were living these behaviours, ensure they were holding themselves to account and to agree actionable improvements.

Result: Embedding and actively monitoring the target culture increased the velocity of pilots going from ideation through to productionisation. Employee engagement within the AI governance team also improved, and the culture to manage risks became more balanced and effective.

4. People and Change Management

Mobilising pilots and proof of concepts (POCs) quickly is important, but true value is unlocked by scaling solutions enterprise-wide and achieving high levels of workforce adoption. Unfortunately, today, 87% of AI products aren’t scaled due to ineffective change management³.

To address this, organisations need to build their change capabilities to navigate the complexities of AI adoption effectively.

From an organisational perspective, it is essential to design end-to-end delivery processes that promote regular performance inspection, continuous learning and fast decision making. From a behavioural perspective, a common pitfall we see is businesses focusing too much on tech solutions rather than on transforming ways of working. AI change is often met with resistance due to fear of job displacement, lack of technological understanding and discomfort with altering established workflows. But business benefits will only be realised if people are willing to adopt new technology and change their approaches for delivering work. This requires effective change management.

Effective change management ensures the workforce is brought on the journey to understand the value opportunity and is supported throughout the change to embed new ways of working. Research suggests for every $1 spent on technology, $5 should be allocated to change management⁴. If you’re investing in the latest AI technology, don’t underestimate the impact of critical business transformation skills and expertise. This oversight is why AI success stories are often slower than expected and cause momentum to be lost.

5. Measure Success

Before launching a pilot, set clear objectives and key results (OKRs). This allows for the determination of whether the hypothesis has been proven and if the pilot was successful.

Measure the velocity of the end-to-end delivery process, from ideation to mobilisation, assessment, and if scaled, production. Fast feedback, continual learning and value delivery should be embedded within this process.

Additionally, track the number of pilots that are scaled, adapted, or stopped. Deciding what not to pursue is as important as focusing on scalable and sustainable initiatives and helps to avoid soaring cloud costs.

Customer satisfaction (CSAT), employee satisfaction (ESAT), operational and financial metrics are also essential measures for assessing customer retention, utilisation, performance and ultimately business value.

Conclusion

Given the rapidly evolving nature of AI, it’s crucial to get started today so not to get left behind.

To begin:

Clearly define what AI success looks like and align teams around these goals
Prioritise use cases that are both high-impact and scalable
Establish a governance framework that balances risk appetite with reward
Invest in change management to ensure the workforce is prepared to be augmented with AI
Define pilot OKRs and AI portfolio metrics to effectively measure success

Following these 5 practical steps will help organisations to optimise their AI journey now, whilst buying time to think more strategically about their longer-term people, data and technology capabilities.

Sources:

To read more on Datagility^TM

Click here