Most businesses have always known things could work better. AI has made that feel more urgent than ever — while doing very little to make it obvious where to actually start.

The reason is rarely a lack of ambition. It's usually the opposite — problems that feel too big to scope, solutions that seem to require turning the whole business upside down, and no one inside with the expertise or confidence to know where to actually begin. The technology being available doesn't help if you can't see a credible path to using it.

And if you've typically relied on agencies, consultancies, or software vendors for your technology, the frustration compounds. You're dependent on people who understand the tools but don't always understand your business — and you end up with solutions that solve the problem they heard, not the problem you actually have.

What I do

Every engagement starts the same way: with a thorough, independent understanding of where the business actually stands. Before any recommendation is made, any tool introduced, or any line of code written, I need to understand how the organisation really operates — its structure, its workflows, its technology landscape, and the gap between how things are supposed to work and how they actually do.

That means talking to people. Not just senior leaders, but the people doing the work day to day — the ones who know exactly where the friction lives, even if they've never had the chance to articulate it to anyone who could act on it. I'm told I have a way of putting people at ease in those conversations: I don't speak in jargon, I don't make people feel out of their depth, and I don't come in with a solution looking for a problem.

What I'm looking for is specificity. Where exactly does work slow down? What gets done twice because systems don't talk to each other? What decisions take longer than they should, and why? And — perhaps most valuably — which parts of the operation are still running on manual processes: things done by hand, in spreadsheets, over email, or through someone's institutional memory, that are time-consuming, error-prone, and quietly fragile. These are often the places where the most straightforward gains are hiding, and where the people doing the work feel the pain most acutely.

The output of that process is a clear, prioritised picture of findings — what's working, what's creating risk, and what deserves attention first — framed in terms of business consequence rather than technical detail. That picture is genuinely useful in its own right, giving leadership the confidence to make informed decisions about where to focus and what to invest in.

For many businesses, that's where they want to start. For those who want to go further, I stay involved to help make it happen — and that's where the next part of how I work comes in.

I don't just advise. I deliver.

Producing a clear, honest assessment of where a business stands — and what it should do next — is genuinely valuable work. But I think what makes that assessment worth more is knowing that the person producing it is willing to stay involved and see it through.

I don't take a templated approach to analysis precisely because I'm not just looking for findings that look good on paper. I'm looking for recommendations I'd be comfortable acting on myself. That changes what you prioritise, how you frame risk, and how realistic your roadmap actually is.

For those who want to go further, I'm not afraid to get my hands dirty. Whether that means building software, introducing tools, working with teams on how to use AI more effectively, or simply being present as changes are made and problems emerge — I stay involved until something meaningful has actually changed. The analysis and the delivery are two sides of the same thing, not two separate services.

Marginal gains, not big bang

One of the things I've seen go wrong repeatedly — in startups and in established businesses — is the instinct to solve everything at once. A sweeping transformation programme, a new platform, a complete overhaul. It sounds ambitious. It rarely works.

The approach that actually moves the needle is different. It's about isolating specific, real pain points and building targeted solutions for them — quickly, without disrupting the rest of the business. Getting something that works into the hands of the people who need it, learning from it, and moving to the next thing. This is how startups win when they're resource-constrained: not by trying to do everything, but by doing the right things in the right order.

It's also how you avoid the paralysis that affects so many organisations when they try to "transform." The typical over-ambition causes procrastination that leaves businesses scratching their heads as to why they can't seem to make improvements that are theoretically available to them.

On AI — a word of honesty

There is a lot of noise around AI right now, and a lot of pressure to be seen to be doing something with it. I understand that pressure. But I don't believe in using AI for its own sake.

The reality is that many of the best automation opportunities in most businesses are still best served by straightforward, well-built software — code that does exactly what it's told, reliably, every time. Where AI earns its place is in the things that genuinely weren't tractable before: understanding unstructured documents, handling variation in language and intent, extracting meaning from messy inputs. That's where the step-change is real.

That includes knowing when the right answer is neither bespoke software nor a complex AI implementation — but simply helping your people get meaningfully better at using the AI tools they already have access to. The best outcome is a team that's more capable and confident, not one that's more reliant on outside expertise.

The same logic applies to every recommendation I make. If a problem is best solved with deterministic code, that's what I'll recommend. If AI genuinely unlocks something that wasn't possible before, I'll show you exactly why and how. You shouldn't have to take that on faith — and with me, you won't.

A bit about me

I've spent over twenty years as a hands-on engineer, architect, and technology leader — most recently as CTO and Co-Founder of Claimer, a VC-backed B2B SaaS platform for tax professionals operating in a regulated, high-trust domain. I built the engineering function from the ground up, acted as de facto product lead alongside the technical role, was actively involved in fundraising, and led regular board-level reporting and communication throughout. I also led the team through a rapid transformation in how it used AI-assisted development, and co-led the business to ISO 27001 certification.

Before that, I led platform architecture and engineering at scale at Gumtree (eBay Classifieds), including a multi-year re-platforming of one of the UK's largest consumer websites, collaborating and communicating regularly with senior leadership at eBay parent company level. I also spent several years as an interim CTO and consultant helping early-stage businesses turn ideas into working products.

What ties all of this together is a consistent pattern: building the right thing, with the right team, at the right pace — without losing sight of the business behind the technology.