How legacy businesses can embrace AI without losing their identity

Business growth consultant Tadhg Guiry gives his three-step approach to integrating AI into legacy companies without losing brand identity.

If done right, AI adoption can drive measurable business impact and productivity gains. Yet, many well-run, successful legacy businesses (despite recognising the potential of AI) still fall into strategic procrastination.

The real question is: why do so many stall when speed is so important?

The answer is rarely about complacency. Instead, I see three main causes why these more traditional mid-market businesses seem to stall when attempting to adopt AI.

Legacy complexity means it’s difficult to identify a clear starting point for AI initiatives. Systems have been built on top of systems over time, and recognising where to begin incorporating AI is a very difficult task.

Also, AI feels multivarious and messy in that it offers many use case possibilities but no clear sequence for implementation. An organisation that hastily tries to introduce AI might face workflow misalignment. This causes those championing AI initiatives to encounter resistance from AI sceptics, creating fragmented and stalled efforts.

Momentum in AI projects often stalls when they are seen as a disruption to ‘business as usual’. On the surface, it feels rational to pause; why risk today’s performance for a future that isn’t guaranteed?

But in reality, those delays are not good forms of strategic procrastination. They’re the kind of caution that leaves a business standing still while the market moves on. And inertia, over time, is far riskier than the disruption you’ve been putting on the long finger.

Challenger versus disrupter

Most legacy businesses see AI as a disrupter, something that replaces people, processes and identity. That framing triggers fear and inertia.

The idea that everything could be replaced overnight is not only untrue, but completely dismissive and misunderstands how these great businesses have survived up to this point –their people.

Rather than framing AI as a disrupter, it’s more strategic to position it as a challenger, one that sharpen – not dismantles – what’s already working.

Legacy businesses don’t need an AI system that tries to replace their people or rip out trusted processes. What they need is something that can constructively question what slows them down: excessive admin, outdated sales processes and wasted travel time. AI should pressure-test these parts of the business, not undermine the ones that hold everything together.

Framing AI as a challenger also helps avoid the ‘outside-in trap’, where adoption happens in silos through disconnected tools or department-level experiments. Instead, it becomes an ‘inside-out’ process, grounded in internal alignment and shaped by the business’s actual structure and priorities.

That’s the core identity piece. These companies have built reputations over decades through service, reliability and expertise. None of that gets thrown out. If anything, it becomes more valuable when the inefficiencies are stripped away.

Seen through that lens, successful AI adoption becomes less about wholesale change and more about smart layering, adding structure in a way that challenges from the inside out.

And when you approach it this way, there are three specific layers that matter most. This is the model we’re using in my own company, and it’s helping legacy clients adopt AI without losing who they are.

Three layers of successful AI adoption

The way I see it, successfully implementing AI with the challenger framework among legacy companies involves building three layers of AI integration into a business.

The first layer consists of setting up AI to verify information across the entire business, connecting core systems (finance, CRM, production, ops) into one accurate, unified view. I like to call this layer the ‘truth layer’, as it affords the organisation an objective perspective of the business as a whole.

Without this verification step, AI won’t be able to function effectively as a challenger because it would end up challenging the wrong things. It’s how you ensure AI-driven improvements reflect your business reality and the promises you make to customers.

The second layer aims to provide the business context. Outside of raw data, it will build an understanding of how the business actually works. Your unique rules, processes, pricing models, customer logic etc. It’s a ‘translation layer’ – it ensures the AI doesn’t operate in a vacuum so it understands the commercial language of your company.

This layer cuts through that by making the system a shared resource. Everyone operates from the same playbook, which turns AI from an isolated initiative into a tool the whole business can use and trust.

The third layer is where AI starts to act, not just advise. Agentic AI can set goals, make decisions and execute tasks autonomously, but only when it has verified and contextualised data to work from. This is where workflows such as margin monitoring, automated proposals or real-time customer follow-ups can run in parallel, without waiting for human input at every step. I call this the ‘execution layer’ – it embeds AI directly into the operations of the firm.

The focus here is augmentation. Let AI handle repeatable, data-heavy processes, so people can focus on higher order problems, while the business benefits from faster cycles, fewer errors and more commercial leverage.

Protecting identity

The fears that legacy businesses have shared with me concerning AI adoption are that AI will dilute the personal service, deep expertise and hard-earned trust these companies have built over decades. These are legitimate concerns, particularly when the approach or strategy for AI integration is short-sighted or misguided. My response to these fears is that AI should amplify these strengths, protecting the business’ identity. AI is most effective when it enables humans to do more of what only humans can do.

Legacy businesses don’t struggle with AI because the opportunity is unclear. They struggle because adoption feels disconnected from how they actually work.

Reframing AI as a challenger with three clear integration steps provides structure, surfaces the truth and challenges inefficiencies without undermining what makes a business unique.

The goal isn’t to become a different company. It’s to become a clearer, faster, more connected version of the one your customers already trust.

By Tadhg Guiry

Tadhg Guiry is the CEO of CSM, where he works with B2B firms to design and execute commercial growth strategies.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.