Insights, articles and news

Is SaaS dead? Not quite. But something has genuinely changed. Vibe coding has made it cheaper and faster to build software than to buy it — and some SaaS subscriptions are already starting to look like an unnecessary cost. But the decline narrative misses something important. The hard part of SaaS was never the features. It was everything wrapped around them: security, compliance, uptime and accountability. Vibe coding hasn't changed any of that. This article sets out what organisations should actually do about it — and why the smartest move isn't replacing everything, but knowing which tools are safe to build, which aren't, and what needs to be in place to make ownership work long term.

Someone in your team has probably already built something with AI. It works. That's exactly why it's worth a closer look. Most SMEs don't have a development team or a CTO — but that hasn't stopped people from building. This article sets out a practical five-step framework for understanding what's already been built, where the risk sits, and when to bring in external expertise to close the gaps.

Sending your data to a third-party AI provider means the risk of accepting decisions you haven't made. Sovereign AI is about taking those decisions back.

Building on a third-party AI model isn't a platform decision, it's a commitment that most organisations won't realise that until they need to move.

Vibe coding doesn't replace technical roles.Vibe coding doesn't replace technical roles. It changes what they're for. This article argues the real challenge is accountability: who governs systems that weren't written by a human, and whether most organisations have ever resourced that capability at all.

Vibe coding has made it possible for anyone to build tools that operational workflows quietly depend on. This article argues the real question is not whether something works, but who owns it, who is accountable for it, and whether governance exists to match the pace of adoption.

Most organisations think they know where their data lives. Ask them to prove it and the picture changes quickly. This article argues that data fragmentation is the single most reliable predictor of whether an AI initiative delivers value or quietly stalls.

AI adoption in mid-market organisations is more likely to redeploy people into higher-value work than replace them, and the real risk for leaders is not job displacement but mismanaging the transition.

AI is often framed as a technical capability, with discussions about skills centred on data scientists and machine learning engineers. This article explores why successful AI adoption depends on a much broader set of organisational capabilities, from data foundations and governance to domain expertise and change management.

In conversations about software reliability, availability targets are often expressed with reassuring simplicity: “We’re aiming for five nines.” Yet behind that short phrase lies one of the most complex, expensive, and nuanced challenges in software engineering.