Why the AI revolution favours your business

AI is levelling the playing field in business, and in some ways, SMEs have advantages over large companies.

That’s the view of César Cernuda, president of NetApp. The San Jose tech company has specialised in data storage, management and security since the 1990s. Its technology is embedded in the cloud products of Microsoft, AWS and Google, and it is valued at around $30bn on Nasdaq. As president, Cernuda focuses on global sales, marketing and customer service, while working closely with CEO George Kurian.

Cernuda took time out on a trip to London to meet me and made a stark point to press his case about the power of SMEs.

Large companies may be better resourced when it comes to their data. They often have a Chief Information Officer and large IT teams. But they typically use only about 30 per cent of their stored data, he says. So 70 per cent lies idle. It could still be useful, but it just hasn’t been structured properly yet.

Getting your data secure and properly organised is the crucial first step if you want to supercharge your business with AI.

While NetApp boasts many large clients, including governments and many of the world’s biggest companies, it also serves plenty of SMEs, often indirectly through service providers, so Cernuda knows this part of the economy well.

Small companies can be more nimble in lots of ways, and data management is one of them. When SMEs begin to work with technology partners, they often have a very clear understanding of their data needs, and they want to be “differentiated” in their approach, says Cernuda. They say: “I am [going to be] embedding technology in a way that other companies don’t.” It can be a more individualistic strategy because they are “closer to the technology”.

Once up and running with their data, the playing field is now more level for SMEs, adds Cernuda. “Small companies, thanks to the internet, have access to a lot of the unstructured data that large companies also have access to, but you can be way more competitive because you can be more agile, you can make faster decisions.” The end goal is to “build models [so] that you can use all that unstructured data with your [own] data”, in a streamlined way.

So what should your strategy be to power up your company with AI? Cernuda says SMEs and large companies have a lot in common. They all want to serve existing customers, win new ones, grow revenue and protect their bottom line.

What also unites companies large and small is the mistaken mindset they can bring to the project.

“I always say, most AI projects fail because we don’t do our homework in the right way. We try to bring our company and its data to AI, and that’s a mistake. What we need to do is bring AI into our company and our data. In other words, I need to prepare my data for AI.”

The first step in any new AI business model is to build so-called “data lakes” to store all the company information you need. This is never easy, though, as you are building it “live”, while using it and adding to it. That has security implications, too.

So another vital piece of advice is to resist the rush to market and the lure of quick competitive advantage.

“You need to make sure your data is protected” in the AI age, says Cernuda. So from the outset, have frank conversations with your data infrastructure providers about how it is protected, who owns it and who owns the models being used with it.

Your data needs to be fit for the data age. Handled the wrong way, it could just as easily become your Achilles heel.

Andrew Moses, CEO of The Config Team, a family-run SAP supply chain specialist based in Cumbria, agrees that SMEs have an advantage when it comes to the AI revolution.

“Each iteration of a small project [in AI] compounds learning,” he points out, “and smaller companies have a higher tolerance for some failures.”

“We have seen that some larger companies have a team focused on AI security, which, although valid, as they have a lot to lose, needs to be balanced with more creativity.”

The key to creativity is to run small pilots, he argues.

“These can be run on subsets of the data pool, without trying to boil the ocean or stifling creativity due to the size of the data pool and the nervousness about its quality.”

“The 70 per cent of unused data stat is the one that should stop every business in its tracks,” adds Jamie King, from Config’s Innovation Team, referring to the data that Cernuda cited. “Companies that are nimble are able to pull apart their data structures with less friction. Who are the nimble companies? The smaller, more agile, light-footed ones! But this advantage only holds if the data is clean, structured and protected.

“Get that right and the playing field doesn’t just level – it tilts.”