How do I set an AI strategy and define a clear business plan for the use of AI?

Working backwards

The most important contribution executives can make to setting up a successful AI strategy is first to clearly define a vision for their organisation and how that vision breaks down into specific, measurable goals. An AI strategy must be aligned with your overarching business objectives. Without this alignment, AI initiatives risk becoming isolated technology projects that fail to deliver tangible value. Start by working backwards from existing business and customer problems and the effect that AI can have on them. The initial phase should focus on envisioning how AI can help accelerate your business outcomes. This means identifying and prioritising transformation opportunities in line with your business objectives. Associate these initiatives with key stakeholders, senior individuals capable of influencing and driving change, and measurable business outcomes. Ensure any adoption strategy is based on tangible, short term and measurable, or at least aspirational, long term and harder to measure, business impact that capitalises on new AI capabilities. A clearly defined business strategy provides data and AI talent with an understanding of what initiatives they should prioritise.

Start with a baseline assessment

Before embarking on an AI journey, it is crucial to understand your organisation's current state of readiness. Assess your AI maturity across various dimensions such as people, processes, technology, and data. This involves looking at questions such as whether you have a clear business plan for AI use, who is assigned to lead AI, how many production AI projects you have completed, and if you possess the necessary skills. This baseline assessment will help identify gaps between your current state and your target state, informing your roadmap and ensuring that your plans are realistic and achievable. Understanding your AI maturity is crucial in determining your readiness to adopt and scale new initiatives. This self-assessment allows you to start small if needed, building capacity and knowledge over time, before committing to larger, more complex projects.

Focus on valuable Use Cases

Once you have a clear vision and an understanding of your maturity, the next step is to identify potential AI use cases. Focus on areas where AI can deliver significant value quickly. Prioritise projects that address pressing challenges or offer substantial benefits, such as cost reduction or revenue growth, to demonstrate AI's return on investment early on. Start with high impact use cases. Good use cases for AI often involve tasks that are repeated frequently, have clear inputs and outputs focused on prediction and optimisation, or involve datasets that can be digitised and structured. It is vital to articulate and quantify the impact and business value of each potential use case, as well as the ease of implementation and associated costs. To achieve meaningful impact, prioritise and map the decisions that will drive the most value. While starting small with pilot projects is wise, also consider larger customer and business problems that can be solved through multiple AI initiatives combined into a hierarchical portfolio. This approach ensures that short term results are shown without sacrificing long term value.

Develop a roadmap

Developing a clear roadmap can help businesses navigate the transformative process of AI adoption. The roadmap should outline a phased approach, starting from pilot projects and moving towards full scale deployment. This plan should be iterative and incremental, allowing for learning and adjustment along the way. Your journey can be structured in four stages: envision, align, launch, and scale. The envision phase focuses on identifying opportunities. The align phase concentrates on foundational capabilities and addressing stakeholder concerns. The launch phase is about delivering pilot initiatives and demonstrating incremental business value. Finally, the scale phase focuses on expanding successful pilots to achieve broad, sustained value across the organisation. This structured journey helps de-risk the adoption process and ensures that every step brings the organisation closer to its goal.

Define needed capabilities

A successful AI strategy is built on a foundation of key capabilities across business, people, governance, platform, security, and operations. Your business plan must account for developing these foundational pillars. This includes creating a data-driven culture, upskilling your workforce, and establishing robust governance frameworks. Data is the key fuel for AI. Your strategy must address data curation, ensuring you can acquire, clean, process, and interact with data effectively. Equally important is the people perspective; building a shared language and mental model for AI across the organisation through fluency training can get buy-in and help business owners adapt. An AI-first culture encourages experimentation and cross-team collaboration. Governance is also critical, helping to orchestrate AI initiatives while maximising benefits and minimising risks, including the responsible use of AI.

Build cross-functional relationships

AI transformation is not solely an IT initiative; it is a new way of thinking for the whole firm. Success depends on cross-functional partnership and collaboration. Strong CIO leadership is essential, but they must partner with leaders across HR, finance, marketing, and other functions to drive initiatives that generate measurable business value. Executive sponsorship is paramount for securing resources, driving organisational change, and ensuring alignment with broader business strategies. A common mistake is to evolve AI units that do not deliver on business value. To avoid this, it is crucial to align the incentives of any AI centre of excellence with your strategy, business, and most importantly, your customers. This collaborative approach breaks down silos and makes AI a shared mission, which is essential for it to take off and reach its full potential.

Define business value and metrics

Defining and measuring success is difficult in AI projects, as setting key performance indicators that are pegged to a business value depends on data availability and model behaviour. From the outset, define what success looks like and establish clear metrics. Articulate and quantify the impact of each potential use case in terms of business value, such as revenue growth, cost savings, or customer satisfaction scores. It is important to have a roadmap that delivers results incrementally over time. Regular monitoring and evaluation of your AI portfolio ensures that your initiatives remain aligned with business goals and continue to deliver value. AI is not a one-time project but a continuous journey of learning and adaptation. Establish feedback loops to continually monitor and improve model performance, and be prepared to adjust your strategy as the technology and your business needs evolve.

Talk to us to explore your options.