Data Integrity - The Boardroom Agenda item that needs to come before AI

"AI Is Only As Smart As The Data We Feed It." - Forbes

2025 has started with a bang - and a renewed focus on AI. The buzz around AI has reached a fever pitch, with breakthroughs in healthcare, finance, and even space exploration, alongside the disruption of the status quo by Deepseek.

The increasing interest and curiosity around AI will continue to grow, however businesses (and people) need to be clear that what underpins the application of AI is data. The reality is that AI applications are only as good as the data they work with. The old adage of “rubbish in, rubbish out”: could not be more relevant when looking at AI. The uncomfortable (and unwelcome) reality for most businesses is that before they seek to embrace the potential of AI, they need to start with their data - is their data ready for AI?

AI is clearly a hot topic for boardroom discussions as companies hurry to increase their knowledge and understanding of what is possible. Generative AI applications, alongside LLMs are dominating the agenda. C-suite roles are evolving, with Chief AI Officers rising in influence and according to a survey by Hays:

• 46% of executives believe the Chief AI Officer will be the most influential C-suite role in the next 2-3 years.
• 90% of leaders plan to upskill in AI within the next 12 months, increasing to 95% in larger organisations.?

“In God we trust, all others bring data.” — W. Edwards Deming

However, amid the excitement, the fundamental issue and challenge of data quality, accuracy and integrity is often being overlooked. If CIOs were responsible for creating Board Meeting agendas maybe it would look at bit like this:

Item 1 - Get our Data Sorted

Item 2 - Go back to 1 until 100% accurate, up-to-date and complete

Then, and only then, spend some serious time on AI strategy

In our recent interview, Dan Mazig and David Walker from Cerve gave a passionate plea to companies. Before investing your time, efforts and money in AI based activities, focus on getting your data spot on. Only then can you realise and harness the clear impact AI can have. Get this wrong, and you will spend more time unwinding errors and problems than getting the benefit

“You can have data without information, but you cannot have information without data.” — Daniel Keys Moran

Cerve, a London-based startup, is on a mission to streamline how the global food and beverage sector shares, processes, and manages data. They believe AI has the power to revolutionize businesses—but only if the data underpinning it is accurate and well-structured.

David urges that whatever new AI programmes are used, “if the data is not accurate in the first place, you're going to have this beautiful system that's going to push through and apply these wonderful processes, but you will ultimately end up with the wrong answer - only faster and with wider ramifications in trying to correct the problems.”.

Cerve uses AI in many of their internal processes and also encounter examples of poorly deployed AI when working with clients and have first-hand experience of when it goes wrong. There are things that AI simply cannot process as a human would, no matter which LLM base is being used. The current configurations, for example, struggle to understand that a pint of milk should be about ￡1. If the underlying data in say a Sales Order Automation tool presents a pint of milk as being 1p, or ￡100, the danger is that AI will take the price, be happy it has found some data, and merrily process away, leaving a wake of errors and admin behind. Without data accuracy and integrity, AI lacks the context to know good from bad, right from wrong, complete and incomplete and as a result, poor data causes AI to generate delays, errors, inefficiencies and problems that can take a long time to resolve.

Cerve’s goal is simple: build a digital infrastructure that delivers accurate, structured, and reliable data. Over 2,000 organizations across the UK and Europe utilise their system including retailers, wholesalers, and manufacturers. Cerve is proving that better data leads to more efficient supply chains, better compliance & product tracking and significantly reduced costs. To solve the biggest challenges in the Global food system everyone needs access to accurate, real-time and complete data. Issues such as food security and global food waste simply cannot be solved without it.

“Data is the new oil.” — Clive Humby (2006!!!)

So in summary, AI can transform your business, but only if your data is an asset, not a liability. Before you commit to investing in AI programmes and applications, ask yourself:

• Do we have structured, accurate, complete and accessible data?
• Is our data standardised from all sources, and in a format we can trust and use easily?
• Have we eliminated the errors, time and costs that come with manual data processing?
• Are we setting up AI for success—or just creating smarter inefficiencies?

If you get your data strategy right first, the opportunities, benefits and impact of the plethora of AI enabled applications can follow.

Written by Helena Houghton, EP Business in Hospitality