Get All Access for $5/mo

Missing the Forest for the Trees: 3 Key Steps to Building a Successful Generative AI Roadmap Leaders should understand the importance of focusing on impactful business processes over individual use cases.

By Ben Bedford Edited by Jason Fell

Opinions expressed by Entrepreneur contributors are their own.

You're reading Entrepreneur Europe, an international franchise of Entrepreneur Media.

Digital43 | Getty Images

Since its launch, ChatGPT and other generative artificial intelligence (AI) tools have been the hottest boardroom topic for their potential impact on business models and markets. In the EU alone, it's estimated over a third of companies have adopted AI. However, CEOs must be wary of shiny object syndrome and resist the urge to chase AI for the sake of it. Leaders who stay focused on what truly matters for their business will get the best out of AI, while those who don't risk missing the forest for the trees.

While it's all well and good to say generative AI will be a game-changer, it's crucial to pinpoint precisely where and how generative AI can have an impact. According to McKinsey, only 12% of businesses sustained their digital transformation goals for more than three years; the risk of wasted efforts with AI transformation could be even higher.

Leaders need to understand the importance of focusing on impactful business processes over individual use cases if they want to benefit fully from generative AI. This means seeing the bigger picture of how the intricacies of a given process could interact with it, and where the technology's limitations lie. Ultimately, commercial value, goals, and KPIs should sit at the core of transformation efforts – and any generative AI use should directly contribute to these objectives.

Here are three key steps business leaders can take to build a successful generative AI roadmap.

1. Prioritise existing metrics.

CEOs have already got a clear view of what happens in their business. They already know what matters to their organization, so AI should support these existing priorities. As such, when approaching AI transformation, leaders should identify – and stick to – metrics that have already been established as priorities.

Getting distracted by AI technologies that don't necessarily align with core business objectives is a recipe for disaster. Too much effort is wasted by people deciding an AI strategy will work across an entire organisation, creating long lists of things AI could do, and then presenting everything back to decision-makers in a menu asking them which options they want. It's a waste of time and money.

Think about some prime generative AI use cases: better summaries of meetings, transcription, and document search can certainly be great pieces of technology. However, this shouldn't consume a company's leadership team and certainly doesn't warrant building a siloed AI strategy based on such individual use cases.

AI is a tool, not a strategy. CEOs would do well to view it as an enabler of core business functions, rather than attempt to create a shiny new strategy with it.

2. Step backward to spring forward.

Next, work back from these priorities to identify potential AI solutions to drive the chosen metrics. To get this right, before even thinking about outsourcing any expertise, CEOs must seek input from their team on whether and how AI can be used to accelerate what they already know to be important.

Take Ocado Group: the British online supermarket has harnessed AI to optimise its supply chain and warehouse operations, which are fundamental to the business. By implementing AI-driven robotics and automation systems, Ocado substantially improved its order fulfilment process, increasing efficiency and reducing operational costs. This strategic use of AI supports Ocado's core business objective of providing timely and accurate grocery deliveries to customers.

That's why one-to-one mapping between the technology and specific metrics and objectives is crucial. Only once the priority process has been identified should it be broken down into individual tasks. Once the priority process has been identified, the next step is establishing which steps should be optimised first.

Here, the best approach is to break down the end-to-end process into a modular series of corresponding tasks and then classify these by their impact on the core business objectives. Categories could include the completion costs (time and money), the predicted economic value added, and the risks associated with inaction (both financial and reputational).

3. Focus on execution – targeting processes over use cases.

Then it's time to execute. This means transforming one essential business process with broad applicability, rather than individual use cases – focusing on end-to-end customer service, for example, rather than just chatbots. By exploring different applications of AI to improve single tasks, AI can then be used to reconnect individual use cases and transform the end-to-end process flow.

Another example: product leads might default to market research analysis as an AI use case – but there's much more on offer. Instead, they could harness AI to streamline the entire product development process, combining customer feedback and competitor research to optimise design. This comprehensive approach ensures that AI integration delivers significant and cohesive improvements across the whole organisation.

Moreover, while the initial priority processes should be tied to one business case, they can provide the foundational technology and capabilities to accelerate future projects, maximising ROI on transformation efforts. Once deployed successfully, business cases for generative AI can be reused in other areas of the organisation to accelerate projects through common capabilities.

Flexibility is also advised. If selected AI use cases prove useful to an organisation's staff but not its core priorities, the CEO should – within reason – let employees use these applications as they wish, as long as appropriate use of AI guidelines are in place.

For best results, CEOs should recognise AI isn't a silver bullet and should be used judiciously. Instead, AI systems should be aligned with existing business goals and implemented incrementally. Once a specific system or tool shows value, C-suite leaders and IT executives can work together to replicate success in other parts of their organisation, connecting each model along the way. Ultimately, business leaders who focus on the forest, not the trees, will put themselves on the path to follow the trajectory of Netflix, not Blockbuster.

Ben Bedford is Head of Delivery & Product Management at London-based Faculty.
Side Hustle

'Hustling Every Day': These Friends Started a Side Hustle With $2,500 Each — It 'Snowballed' to Over $500,000 and Became a Multimillion-Dollar Brand

Paris Emily Nicholson and Saskia Teje Jenkins had a 2020 brainstorm session that led to a lucrative business.

Business Ideas

63 Small Business Ideas to Start in 2024

We put together a list of the best, most profitable small business ideas for entrepreneurs to pursue in 2024.

Business News

'I'm Not Trying to Land on Mars': Mark Cuban Takes Dig at Elon Musk to Explain Why His Online Pharmacy Isn't Trying to Make More Money

Mark Cuban Cost Plus Drug Co. is an online pharmacy co-founded by Cuban and radiologist Alex Oshmyansky.

Business News

'It's Not About You': How to Fire Someone Effectively, According to Kevin O'Leary

O'Leary says that if you can't fire someone, you aren't the right leader for the organization.

Business News

Meta Makes $1 Million Dollar Donation to Donald Trump's Inaugural Fund

Meta CEO Mark Zuckerberg also reportedly gave Trump a pair of Ray-Ban Meta smart glasses.

Leadership

Should I Stay or Should I Go? 8 Key Points to Navigate the Founder's Dilemma

Here are eight key signs that help founders determine whether to persevere or let go.