AI Is Not a Quick Fix Tom Allen on smarter integration, marketing innovation and long-term ROI

As one of the UK's leading artificial intelligence (AI) speakers, Tom Allen, founder of The AI Journal, a contributor led platform with over 2,000 thought leaders from some of the biggest institutions, tech vendors, and enterprises, is on a mission to demystify AI for business leaders. Drawing on a background in brand strategy and experience advising fast-scaling tech firms, Tom helps organisations understand where AI truly adds value - and where the hype falls short.

In this exclusive interview with BIZ Experiences UK, he explores how companies can realistically integrate AI into their operations, what marketing teams are getting right (and wrong), and the biggest misconceptions decision-makers still have about AI adoption.

Before integrating AI into a business, what foundational steps should leaders take to ensure it delivers long-term value?
Before you look at the integration of AI into the business, you need to look at how you're structuring your business. And this is what people get normally, from what I understand, get backwards. It's more like an assistance tool. So, to integrate it, you've really got to think - and there's plenty of experts in this area - for understanding how you want to transform your business. That usually starts by just very small tweaks - by understanding what is it you're looking to do better, looking to do faster. What is it that's going to be beneficial either for the customer or beneficial for your time and usage?

There's a lot of tools. If you use Google Workspace or if you use any form of large language model, they're building it into workflows. And a lot of SaaS companies such as Salesforce, Workday, Monday, Notion, they're building it in. It's actually quite easy to integrate into your day-to-day workflows because any big vendor that you're buying from probably has that capability. But it's normally a step-by-step process of thinking: where are the gaps, what do we need to do faster, what's being required from the customer - because people expect it more these days. There are two sides: firstly on the talent side, but also understanding what your budget is for it and understanding how does it fit with what you deliver as a business, as a service for your customers.

I think that's where a lot of people go wrong. It's very easy to get caught by the flashy tools and new plugins and new benefits, but it's always quite key to focus on how do you do business - because it's still very much relationship-focused - and whatever can make that better usually is the right tool to go with, or the right upskilling to go with.

What are some of the most practical ways AI is being leveraged in marketing today - from content creation to campaign execution?
Anything from content creation to analytics to building templates. Look at Canva - they've built it in so you can use that from imagery to designing websites. You can do coding or get platforms to build it for you, like Replit or Lovable or Bolt. All of those are the face of your brand - the communication side. What you need to do in terms of showing what your brand is to a customer, but also just having a lot more fun.

There's video generation tools where you can create campaigns and create ten different variations very quickly. Think about how hard it was - I used to work in marketing just five years ago - to come up with a campaign. You'd have to really think and iron it all out, and now you can just tell a large language model. I'd imagine agencies will be building their own models that are in tune with previous data, and they can do anything from testing ad copy to testing entire multi-million-pound campaigns for how it will resonate.

And then you can analyse the performance. That's a big reason why people use Meta ads and places like that, because you can analyse performance on cost per lead and all those kind of things that marketing people get very excited about. It depends -how an energy company markets themselves is going to be very different to how an energy drink company markets themselves. How a Red Bull markets themselves compared to how a day-to-day living cost or a Tesco markets themselves will be very different. How you use those tools can be much more in line with who your customer is - connecting with them better. And that's essentially the key to marketing: to connect to your customer better.

In your experience, what's the biggest misconception among executives when it comes to AI implementation and ROI?
How quickly it can get you a return on your investment. I look at it from the perspective of how we use the internet. I'd say only really recently we've been using the internet at the high capacity we have, and we're going to use it more and more. We want things faster. Our attention span is less. With AI, the model is going to get better. The data is going to get better. People that aren't so tech-savvy are going to learn it better.

People will spend a lot of money on these tools or with vendors expecting high returns, but it's like trying to get your business to double its revenue every year. It takes time. You can't wake up one morning and just find everything fixed. I think people misunderstand that - with those fixes come problems. Even in the big companies like Google, when they release new models, they have a lot of backlash on it, and the model gets better because of that. And I think that's a very positive thing to have, because you need to improve the model. I think people going into it thinking that they are going to get return on their spend or it's going to not create any problems is daft.

You look at any kind of industry - whether it's travel, sport, entertainment, media or construction - all these areas, there's a lot of applications. But to get people using it day in, day out - the adoption - it takes a long time. And to use those in sync and to get your data in place, you're going to have to go pay either some very good developer or you're going to have to use a company like Accenture to understand how you assess all what you've got and then try and make sense of it to apply an AI layer over your business.

The people that are not doing it today will be the ones that will lose out in five years. But it's very hard to take your money and put it into something you might not see a return on in five years. Fortunately, most people that run businesses can see the vision for five, ten years from now.