AI strategy
Work out where AI is worth using, what it could change, and which risks matter — then leave with a practical, prioritised roadmap.
Explore AI strategyIndependent AI consulting for UK founders, SMEs and established businesses. I help you decide where AI is worth using, then design, build and hand over something that works.
Based in Yorkshire, working with UK businesses remotely.
I work directly with clients, not through an account team. The work is shaped around the problem, not a standard AI package.
That might be a short strategy engagement, a hands-on integration, an automation project or senior engineering support. The aim is the same: make the business work better.
Work out where AI is worth using, what it could change, and which risks matter — then leave with a practical, prioritised roadmap.
Explore AI strategyAdd LLM features, knowledge assistants, or chat and voice agents to the systems you already use, with clear human handover.
Explore integrationsTake repetitive admin out of enquiries, follow-up, reporting and data transfer, while keeping people in charge of the judgement calls.
Explore automationArchitecture, legacy modernisation and AI-assisted development, with sensible testing and a codebase your team can live with.
Explore engineering
I’m Mike Atkinson. I’ve spent 25+ years building software businesses, and built and sold two companies. I founded e-GoodManners, used by 1,500+ dealers including VW Group UK and Toyota/Lexus, and built systems for VW Financial Services and Toyota Financial Services. Companies and interests of mine have been acquired by Cox Automotive and CITNOW Group.
I still build. Carro Systems, Leadtrak.ai and WasteTrak.ai are live products, not case-study fiction. That mix of founder experience and hands-on engineering is what I bring to consulting: an understanding of the commercial problem, and the ability to make the technology work.
These are live products I have built or co-own, not consulting-client testimonials. They show the kind of problems I like: messy processes, disconnected systems and a need for something practical.
Cloud-based dealer CRM bringing enquiries, stock, diary and tasks, quotes and order forms, and reporting together. The work includes integrations and AI enquiry and chat capabilities with human handover.
Read the Carro Systems studyLead capture connected to qualification and source attribution, with mobile-first funnels, lead scoring, automated follow-up and cost-per-qualified-lead reporting.
Read the Leadtrak.ai studyAn AI waste-photo analysis platform I co-own, with mobile and web access, waste composition reports and REST API integrations. Finished and live, not a proof of concept.
Read the WasteTrak.ai studyNo mystery, no long discovery theatre. Four straightforward stages, with the scope agreed and the working system tested together.
Start with the work, not the model. We look at where time, knowledge or follow-up is getting lost.
Define a small, useful piece of work with a clear result and an honest view of the risks.
Short loops, real examples and human review at the points where judgement matters.
Documentation, practical training and a system your team can run without me.
A few questions I’m asked before a first conversation. If yours is not here, ask it directly.
An AI consultant helps you decide where AI is genuinely useful, then plans and builds the work needed to put it into practice. My focus is practical: understand the problem, assess value and risk, and build something your business can use and maintain.
Yes. I work with UK founders, SMEs and established businesses. The work is sized to the business - sometimes a short strategy piece, sometimes a hands-on integration or automation project.
Yes, where the systems expose suitable APIs or other safe ways to connect. I have integrated CRMs, finance systems, websites and third-party services, and I will flag clearly if an integration is not sensible.
Start with a process that is frequent, costly in time and has a clear owner. A short opportunity and risk assessment is usually the best first step, so you can choose one useful piece of work rather than starting with a tool.
No. AI can reduce effort and improve consistency, but the result depends on the process, data and how well the system is adopted. I aim for measurable improvements and honest limits, not guarantees.
Tell me what you want to improve. I’ll tell you honestly whether AI is the right answer, and what a sensible first step looks like.