How Pets at Home Bucks the Trend of AI Inaction in Europe

The vast majority of the economic value of generative AI (GenAI) is expected to come from its adoption by European organisations, according to McKinsey & Company. They estimate GenAI could add $571 billion to the European economy by 2030.

Yet, while the opportunities are wide open, many businesses in Europe are still moving slowly compared to other regions, as shown by Deloitte research.  

This hesitancy means a lot of investment (and innovation) in AI is happening elsewhere. Since 2022, more than 90% of LLM-related funding has taken place outside of Europe, according to the Stanford University AI Index. And the European Court of Justice’s October 2024 ruling could further harm progress, with lawyers saying it’ll “serve to slow down technological innovation in the EU.”

One European company, however, that has taken the leap is Pets at Home, with its prioritisation of AI in retail. We spoke to Simon Ellis, Head of AI Transformation and Enterprise Architecture at Pets at Home, about the steps other businesses should be taking in their AI development.

1. Navigating today’s AI landscape   

While there is so much hype about AI, it’s hard to know what’s possible and what’s bluff. And because AI is developing so quickly, for some organisations paralysis sets in. So, Simon says the first thing companies can do is remove the fear by not expecting immediate results.

For effective experimentation, you need access to data. Simon says, “Of course, gathering the data and knowledge needed to train your AI models is key. And getting clean, quality, structured data that’s in the right kind of language and tone that you want, can be a challenge.”  

Once off the starting line, knowledge about what’s possible with AI increases rapidly. Simon says, “A big advantage of prototyping is that you get to learn more about the quality of the knowledge you have access to – and it doesn’t need to be perfect to get started.”  

What can you do to start experimenting with AI now?   

• Rather than broadly applying AI to your business, think about specific use cases.  
• Hire specialist roles that understand the technology and can think strategically about using AI in your business.  
• Don’t wait for the perfect moment or dataset; get started now.  

2. Democratising the use of AI    

Another issue potentially holding back GenAI in Europe could be its limited adoption among the general public. Salesforce research shows that 29% of the UK population uses GenAI compared to 45% in the US and 73% in India.  

AI penetration needs to improve, especially if businesses are to take advantage of the next generation of tooling, which promises to democratise AI using low-code systems. How can companies empower non-technical users to experiment with AI in their workflows?   

For Simon, the answer lies in taking a very personalised approach to what AI can achieve. “Business users can use AI platforms to streamline and innovate their individual processes – not on a production level, but on a case-by-case basis.” The prototypes built using natural language models should start to make life simpler too, as “the need for complex engineering on AI initiatives starts to reduce – although not necessarily eliminated,” says Simon.    

What can you do to democratise AI now? 

• Don’t throw employees in at the deep end; allow them time to become familiar (and experiment) with today’s GenAI tools.  
• Change is hard for a workforce; engage employees with the process by inviting suggestions for AI-enabled improvements.   
• A little training goes a long way. Employing specialist educators will help employees unlock the full capabilities of GenAI.  

3. Managing data governance and policy   

It would be easy to point to EU regulations, such as GDPR, to explain European organisations’ reticence to experiment with AI. Simon says, “I think it’s fair to say that we all have our eyes on the legislation around AI … we’re watching that carefully.”    

However, these regulations are intended to foster responsible innovation by ensuring privacy and ethical standards. Rather than seeing them as inhibitors, businesses could see them as useful guardrails that enable safe AI practice.   

Simon leans into legislation. “We’ve taken the approach of innovating consciously alongside our legal and data protection teams to explore and … to set the policies and standards for the organisation.” The result of this collaborative approach is clarity around how they should safely utilise AI at Pets at Home, including internal compliance training and data security protocols.    

Most importantly for Simon, this approach “doesn’t stop us from experimenting, innovating, and prototyping.”  

 What can you do to ensure compliant AI practice now? 

• Work with legal from the outset. They’re on your side to make AI work, legally.   
• Reframe regulations from a barrier to your creative framework for innovation.  
• Make AI adoption a collaborative task that involves every area of the business.  

Do it for the customers  

While there are many factors causing European businesses to drag their feet over AI adoption, Deloitte research shows that nearly all CX leaders are confident that AI has the potential to improve customer experience. But these opportunities won’t come to fruition unless companies have the confidence to invest in and experiment with AI. 

For those still too cautious, Simon’s final piece of advice is a simple one: “It’s important to try new things.”