Want to generate value from AI? Invest in a top-class data strategy, then have a little patience.

2/04/2025

How can senior leaders truly prove the value of a technology as new and ‘unproven’ as generative AI? This was the core question posed at The Economist’s 5th Annual Business Summit: AI to ROI earlier this week.

Being huge advocates of AI and the opportunities it holds for businesses, Grayce were proud supporters of the event. The day set out to empower business leaders to innovate and seize growth opportunities with AI and other emerging technologies via talks from a suite of great minds in this area.

Of course, many of the insights will remain with those in the room, however, we don’t want to keep all the learnings to ourselves, so we’re sharing some of our key insights from the day for your reading pleasure. Grab yourself a cuppa and let’s get going.

Why good data matters

As the saying goes, “rubbish in, rubbish out”, and that’s never truer than when considering the data we’re feeding AI. Jessica Hall, chief product officer at JustEat quite rightly communicated that a tight data strategy is essential for effectively leveraging AI within an organisation. Gen AI is only ever as good as its data strategy, which is why availability of accessible, high-quality, reliable, timely and clean data is key. While most businesses are data-driven in this day and age, many haven’t invested in a proper data strategy; this is an essential focus if businesses want to deliver true value through AI.

Agentic AI is taking off

When you visit any industry event, regardless of the sector, there tends to be a core theme or trend that comes across most strongly, discussed on various panels and overheard in discussions amongst guests. The phrase on everyone’s lips at the Business Innovation Summit: agentic AI. This refers to AI that’s capable of independently pursuing goals through reasoning, planning and autonomous action, freeing up humans for higher-level thinking and creativity. This was cited by many as the next step on the hype curve and possibly the differentiator that starts making us start viewing AI ‘as a teammate, not a tool’ – we took that phrase from an Economist interviewer, but we like it.

A tech change is a people change

Generally speaking, leadership is bought into AI. This point is proven in our latest research which found that 2 in 5 FTSE350 companies have a dedicated head of AI/lead.

However, buy-in at the top doesn’t always mean buy-in elsewhere - particularly when that change is related to something as technical and vast as AI – so it must be managed properly. According to Jessica Hall, when it comes to AI implementation, introduce clear rules around the use of AI and new tech; make sure people understand the policies clearly but can still use the tools. If you ‘block’ AI completely within your organisation, colleagues are likely to use the open tools regardless, leading to data and security risks. It is best to take control, mitigate risk, and steer AI’s use, whilst empowering employees.

Encouraging curiosity

Those curious and open to learning are more likely to be bought in long-term. With that in mind, Kellanova has implemented organisation-wide ‘Kuriostiy clinics’ so staff can get curious about AI and new tech.

Chief data and analytics officer, Loretta Franks, explained that these regular sessions are led by IT and attended by colleagues at various levels throughout the organisation, including leadership. They set out to encourage interest in AI and provide a space for people to be innovative, using experts within the business to educate upwards. We love this idea, sharing values with our reverse mentoring initiative here at Grayce.

A shift in mindset on skills

It’s easy to think that the core skills required to make an AI implementation effective would be technical; however, that’s not necessarily the case. Bastien Parizot, SVP of IT, digital and transformation at consumer giant Reckitt advocated for prioritising soft skills over technical skills, citing learning agility, adaptability, and resilience as the most important skills.

We’re in agreement here at Grayce, prioritising the ‘8 skills of highly effective Analysts’, (including resilience, adaptability, proactivity, problem-solving and communication amongst other soft-skills), when recruiting our Analysts - knowing that technical skills can be taught. Ultimately, AI is moving and changing at pace, so associated technical know-how is evolving with it.

Jump in but bide your time for ROI

Many panellists made reference to the importance of patience when it comes to evaluating AI’s ROI. Businesses should give AI tools time and space to bed into the organisation and be tested and tweaked for a few months before they seek the ROI.

Remember that change moves at the pace of business, not technology. Naturally, it will take time for organisations and individuals to learn how to use this revolutionary tech most efficiently and effectively, so time is essential when it comes to ROI. That being said, a core sentiment was that although we should wait to chase results, we must adopt AI now; organisations that start using AI tools today are already behind, and the pace of change will only get faster.

Closing thoughts

We’ll finish on a sentiment from Aneesh Raman, chief economic opportunity officer at LinkedIn, who cited a quote he’d heard at a recent event “The {technological} change will never be as ‘slow’ as it is right now.” Change will only come thicker and faster, so companies need to be leaning into AI now to avoid being left behind.

Photos by Economist Impact