Highlights
- If 2026 was about AI experimentation, 2027 will be about measuring return on investment and proving its worth.
- For mid-market companies, that means tying technology decisions more closely to business architecture and economic outcomes.
- Embedded across the enterprise, AI can help drive efficiency gains, devise new products and enhance existing ones, and even reduce technology debt.
The Annual Inflexion Product and Technology Exchange saw business leaders discuss their expectations for the year ahead, with sentiment shifting from “where can we use AI?” in 2026 to “where can AI genuinely improve growth, resilience, customer outcomes or cost – and how do we know?”
“It’s been a year of learning and discovery,” says Mike Arshinskiy, Managing Director in the Value Acceleration team at Inflexion, opening the 2026 Inflexion Technology and Product Exchange, hosted in collaboration with AWS.
The brave frontrunners reap benefits, with companies that have invested heavily seeing clear upside. “A lot of AI is landing and we’ve encouraged our portfolio to experiment. We see early results and it is time to industrialise the approach across all our portfolio companies,” he says.
AI is a cross-functional issue
Engineering teams want to build products that are useful and widely adopted, but their roles are shifting fast with AI. The sentiment around it is mixed: according to third-party research surveying engineers, 41% were energised “all-in” adopters, 35% felt conflicted, 12% were disoriented by the constant change and 12% resented the pressure to adapt
It means adoption is an organisational challenge as well as a technical one, and so businesses need to support teams through the transition rather than treat AI as another technology rollout.
And while AI has raised productivity, it’s important not to fall behind as it’s also raised competition. Mike explains B2B AI adoption is a chance to get ahead, so he advised budgeting for AI spend the way you budget for infrastructure, and help your team through the change.
Redesign before you automate
AI should be applied carefully to existing workflows. “Beware blindly following processes. Automating a process that shouldn’t exist makes no sense,” warned Phil Le-Brun, Executive in Residence at AWS. He suggests looking to improve processes – or radically rethinking them – and focusing on the outcome rather than the product itself.
Phil is co-author of The Octopus Organisation, a management approach that swaps rigid corporate structures for more adaptable ones, where decentralised teams make local decisions but stay connected to a central brain. He says people need clarity of purpose and problems to solve, and want ownership, and that organisations need to tap into their curiosity.
From experimentation to measurable impact
Reviewing 2026 from an enterprise IT perspective, Phillip Dixon, Assistant Director of Enterprise IT at Inflexion, also saw a gap between experiment and result. AI has delivered visible task-level productivity gains, but turning those into measurable P&L improvement is harder. Nor has it yet disrupted traditional SaaS as much as many predicted.
He expects that to change. The ahead should bring more innovation, faster delivery, stronger resilience and lower costs, with AI initiatives increasingly absorbing legacy technology debt. Managing the transition across complex enterprise technology estates will be the challenge. “Invest time and effort into enterprise architecture. It will be crucial to success in technology platforms,” Phillip says.
Experience is changing in value too. AI is making individual technical skills cheaper and easier to come by, which raises the relative importance of people who understand the business, its customers, its systems and the consequences of different choices. CFOs will expect more as well: 2027 is likely to be when finance teams ask for a clear line between AI spend and the P&L.
People plus AI
None of this means replacing people. John Sullivan, Group IT Director at DWF, argued the strongest model combines what each does best: AI brings speed, scale and pattern recognition; people bring empathy, judgement, creativity and leadership. The winners will learn the tools by using them on real work, within sensible guardrails, while building human skills like communication and collaboration.
Mike concurred with a widespread view of the day’s speakers. “Critical thinking is the strongest resource you can have in your team and this will be crucial next year,” he says.
How does Inflexion help portfolio companies capture the AI opportunity?
AI cuts across data, systems, cybersecurity, commercial strategy, talent, M&A and governance, so Inflexion’s dedicated value acceleration team works alongside management teams across all of these areas. “It is very much a cross-disciplinary effort rather than simply a technology challenge,” says Alex Mathers, Inflexion Partner in charge of value acceleration.
The specialist AI and data capabilities required are scarce, and many mid-market businesses have not previously needed to build this kind of expertise in-house. “We help strengthen leadership teams by recruiting specialist talent, coaching existing executives and addressing knowledge gaps, drawing on Inflexion Ignite, our global network of trusted value creation specialists,” Alex explains.
Inflexion also work closely with major technology providers and implementation partners, helping portfolio companies access expertise and commercial advantages that can otherwise be difficult for a mid-market business to secure independently.
