A recent ODI webinar debated the needs for boards to engage with data and the digital revolution.

We talk about the need for data literacy in the finance function, but what about the boardroom? This was the main thread of discussion on the Open Data Insitute’s podcast, Why Data Literacy is Needed in the Boardroom.

On the panel was Generation CFO’s founder Chris Argent, along with Dame Wendy Hall, Regius professor of Computer Science, University of Southampton, Abubakar Suleiman, CEO, Sterling Bank, and Helen Crooks, Chief Data Officer, Ofgem and Ambassador for AI and Data Economy.

The pandemic has brought to the fore how important it is to understand and use data, said Dame Hall. “We need more expertise on the board, not just reporting into the board.”

“I don’t see a good level of data literacy at board level,” said Argent. “We need to be honest and open about the talent coming through, there’s a friction between this talent and the board.”

There is little understanding of the opportunities that AI offers among board members, and something needs to be done to bridge that gap. “The first rule is to equip the board to build a vision for the company,” said Suleiman. “They need to be able to understand what is being presented. Then we need to have a deeper understanding of data in the boardroom. Bringing people in who have this knowledge is a useful start, but the end game needs to be that everyone in the room needs to understand these conversations.”

How can that process be accelerated? As everyone starts to use more AI, things will get more onerous, said Dame Hall. “There are going to be a lot more regulations on how you use algorithms and how you represent the data. Are the algorithms introducing bias? Can people trust what your company is doing?”

In the financial world, there are rules that everyone has to adhere to; that level of regulation will come through for the rest of the business as well. “I think we’ll have a legal requirement for an ‘AI data person’ on the board,” she said. “It will happen quite fast. The EU is looking at more regulation now and I see it happening all around the world.”

Interested in more articles on data literacy? We recommend these.

The lack of regulation is in a way holding back the data discipline, so its impacts are not always understood at board level, said Crooks. “People seem to want to wrap it up in jargon and lingo. I believe the board needs to get a grip on that and present it in a human way.”

“It’s using plain english,” said Argent. “There aren’t many standard definitions in this area but there is a general language we can follow. One of the biggest failings of the data community is not being able to present this in business terms. A lot of people still see this as highly technical, and IT-led, but actually it’s not. To get your data managed and well governed is a business asset.”

Three things that are important when it comes to data and governance, said Suleiman. First, compliance: what are the sources of liability and risk. Secondly, data as an asset. “We need to think not just of data in the business but data as the business. This is the best way to speak to the board, when they can see the economic value of data. Give it another couple of years and we’ll be able to see data presented on the balance sheets of the business.”

The last is how to reward data. “The board struggles to reward data. If you don’t find that alignment between talent and data assets that you are creating, the risk is you undercompensate valuable talent and you lose them.”

Argent mentions Arkadia, which recently called in the administrators after months of financial struggle. The business that didn’t have data at its heart, which was one of its biggest failings. “Talking to the management team there five or six years ago, there was very little appetite for that data journey. If you start now, you can build this out over years, but you do have to start. Doing nothing can be a problem.”

There is clearly a difference in strategy between Pretty Little Thing and TopShop, he said. One has got a Tik Tok account and millions of social media followers. Therefore, it has millions of data points that can be used for reporting, planning and data management. “The other one has nothing in that space.”

How do business develop and grow talent? “Come and take one of our courses,” said Dame Hall. “That’s easy to say, but it’s difficult to get executives to do it. Everyone knows you need that knowledge, but for the most part, you aren’t breaking the law if you don’t do it. That will change when more regulation comes in.

A board that doesn’t use AI and data to move forward won’t be a board for much longer, the panel agreed. Companies have to start investing in their own training programmes. “You get lots of different types of data officers or data capabilities coming up through your organisation,” said Crooks. “Trying to find the right individuals to help create that board level strategy is imperative.”

Listen to the full podcast here.