Executives Use 3 Strategies to Build Strong Data Foundations for AI Integration

Business leaders are well aware that robust foundations are crucial when it comes to harnessing the power of artificial intelligence (AI). Diving into AI projects without a solid data strategy in place is a recipe for disaster. As the old saying goes, "garbage in, garbage out"—and that couldn't be more true when it comes to AI.

So, how can professionals lay the groundwork to ensure their organization can use AI both safely and effectively? Here, three business leaders share their top tips for crafting a successful strategy for leveraging emerging technology.

1. Put Your People First

Claire Thompson, the group chief data and analytics officer at insurance giant L&G, stresses that a strategic approach to information is vital for any company looking to innovate. "I always say data foundations are important for whatever you do next," she told ZDNET. Thompson emphasizes the need to connect rules and regulations to tangible financial outcomes.

"Make it clear how the data strategy will drive tangible value—why is it important, for example, that your email addresses are up to date and accurate so that you can do targeted digital communications?" she asks. Thompson understands that not everyone is thrilled about diving into a long-term strategic plan that outlines the technology, processes, people, and rules required to manage information assets. However, she insists that the planning stage is critical to reaping the benefits of technologies like AI.

"I can understand why people might say governance is boring," she admits. "But in today's digital organizations, where people want to do straight-through processing, it becomes even more critical that your data is good quality. So, all roads are leading to governance."

A key part of Thompson's strategy at L&G is fostering a close working relationship between her data team and the IT department. Effective collaboration hinges on clarity about the skills each party brings to the table. "You need a hand-in-glove partnership. Technology is hugely important to what we do in the data space, and we can't do our work without the cloud environments, the data warehousing, and the tooling. Data is held in all the applications that the IT team maintains," she explains.

Thompson also highlights the importance of embedding data quality by design into core systems. "The more you can do that work, the more it stops the ripple effect of poor data quality further down the line and prevents any remediation effort," she says. This approach will pave the way for enhanced customer experiences, such as personalized mobile applications and automated trades supported by AI.

2. Master Your Transactional Data

Jon Grainger, CTO at the legal firm DWF, believes that now is the perfect time to develop a data strategy. He stresses that savvy business leaders should focus on the foundational elements of data use well before considering how to leverage AI and machine learning. "I always say the best time for a data strategy is four years ago," he quips. "It's a supertanker piece of work. Ultimately, there aren't many shortcuts. There is a view that says, 'Well, if it's going to take that long, why bother?' And I think that's why many folks haven't been able to get to grips with their data."

Grainger's goal is to help his firm build a reputation for delivering great experiences through digital transformation—a data strategy is a crucial component of this approach. Since joining DWF in late 2022, he has implemented a new strategy centered on cloud-based software-as-a-service (SaaS) products and open application programming (API) interfaces.

The data at DWF covers various entities, such as cases, partners, clients, and internal business processes, including billing and financials. "The data strategy is all about ensuring transactional data—the source of truth—is mastered in those sections," Grainger explains. The aim is to help the organization move quickly without compromising quality or cost.

"Each SaaS product has a clear identity on the enterprise map," he says, detailing the nuances of his data strategy. "That identity is driven by the data you master in each area." Grainger emphasizes that the "absolute minimum requirement" to integrate into the firm's target architecture is well-developed APIs that DWF can access and use.

He points out that SnapLogic technology plays a key role in ensuring a solid and reliable connection between services, APIs, and users. "Invariably, you'll get 15 different spellings of a particular address, and the technology can see that pattern and correct it," he says. "It can also do something called enrichment. So it might take someone's reference, go off to an API, come back, and say, 'This is the right information.'"

Grainger also notes that the data strategy focuses on the models DWF creates to answer key business questions. By combining this with the firm's focus on SaaS products and APIs, DWF has solid foundations to explore emerging technology. "It turns out you're setting yourself up pretty well for generative AI if you've got all those elements in your data strategy," he concludes.

3. Work with Your Industry Peers

Nic Granger, director of corporate and CFO at North Sea Transition Authority (NSTA), believes that a great data strategy extends beyond internal practices and spans organizational boundaries. NSTA collects data from the oil and gas sector, and Granger's team has developed digital platforms that allow industry, government, academia, or other interested parties to access data openly.

As part of this effort, she chairs the Offshore Energy Digital Strategy Group (DSG), a specialist body formed in late 2022 to foster collaboration across UK public bodies involved in data collection in oil, gas, and renewables. "It was recognized that we needed a cohesive digital data strategy across the offshore energy sector," she told ZDNET. "There were good pockets of excellence across the industry in data management and digital technologies, but they weren't necessarily talking together. So that was a big priority for us."

The DSG is supported by contributors including the UK government departments, the Open Data Institute, and the Technology Leadership Board. Granger says this collaborative approach has been fruitful: "We've got the data strategy now, and it's about working on three key streams of work."

The first stream focuses on data, standards, and principles: "Making sure the underlying quality of the data is good because we're all working on the same basis," she explains. The second stream aims to create common data toolkits and ensure interoperability. "It shouldn't matter if you're working in an offshore energy company or on a project in an oil and gas company, you should have data that's useable across the platforms. That work is all about, 'How do you get that data from A to B without duplication?'"

The third workstream focuses on cross-sector digitalization: "That's about ensuring the data and digital skills are there across the industry, and ensuring the sector complies with cybersecurity best practice," Granger says.

With these data foundations in place, it's much easier to start thinking about how to make the most of emerging technologies. "Our focus is on ensuring we're making the data accessible and in the right formats for others to use AI and machine learning," Granger concludes.