Prepare
Finding information, cleaning it up and copying it into the tool.
AI Strategy & Implementation Review · UK SMEs
AI tools in place.
Work still piling up?
Your team is copying between systems, checking AI answers or keeping a pilot alive by hand. Find what is blocking progress and agree what to fix, connect, train or stop.
Free 20-minute conversation with Justin Sherwood. For SMEs across all sectors, from first experiments to existing AI deployments.
Which problem
sounds familiar?
The useful question is where work breaks down. Select a problem to see what we would examine and the decision the review should help you make.
Choose a setup the business can run.
A local model may fit your information-handling or connectivity needs, but speed, quality, integration and support still need to work.
- What quality, document length and number of users must it support?
- Which parts of the workflow actually stay local?
- Who owns the hardware, updates and recovery?
A useful outcome: A local/cloud/hybrid suitability recommendation and a separately scoped evaluation plan where testing is needed.
Illustrative starting points. Your selection stays in this page and is not an assessment of your business.
Follow the whole task
Count the work around the AI.
Generate
The part handled by AI: a draft, summary, extraction or suggestion.
Check
Correcting mistakes, resolving missing information and approving the result.
Complete
Entering the result into your systems and making sure someone acts on it.
Measure the full task, including exceptions, before deciding whether a change has helped.
Local, cloud or a mixture?
Local AI still needs
an implementation plan.
Running a model on your own computer or server may suit a workflow with particular data or connectivity needs. The choice also needs to work for your team, systems and budget.
It is too slow for everyday work.
Plan a representative workload test: document length, quality and simultaneous users. Compare hardware capacity and a suitably sized model before buying equipment.
The answers are not good enough.
Use the same task examples and acceptance criteria for each option. Better source material, a narrower task or another model may help; local hosting alone does not improve accuracy.
We need to control where data goes.
Map the whole route: model, documents, connectors, logging and any cloud fallback. Define access and permitted data flows; a local model does not make every connected feature local.
It only works on one person’s laptop.
Assess shared service hosting, permissions, workflow integration and support ownership. A desktop demonstration needs a plan before it becomes a team dependency.
Nobody owns updates or recovery.
Define responsibility for versions, patching, monitoring and recovery. Test changes on agreed examples before rollout and keep a workable manual fallback.
The model is free. Is the service cheaper?
Compare equipment, energy, setup, support and team effort with the cloud alternative. Check the selected model’s terms. The right answer depends on actual workload and usage.
The review includes a local/cloud/hybrid suitability recommendation for the selected workflow. Hardware purchase, installation, benchmarking and production deployment are separately scoped.
The review
From a frustrating pilot
to a clear decision.
£1,495 fixed review fee*
One SME team or business function. Up to five candidate workflows and three existing AI tools or pilots. A focused review of both your direction and the practical work needed to implement it.
- Show us the real task.A 90-minute workshop with up to four colleagues. Identify the business goal, workload, systems, recurring frustrations and who owns the work.
- Find the implementation gaps.A 60-minute workflow or tool walkthrough and up to ten redacted examples. Examine manual handoffs, output quality, usable information, access, staff adoption and costs.
- Decide what is worth doing.A written gap review and prioritised shortlist, including local, cloud or hybrid suitability for the selected workflow. Each recommendation states what to fix, connect, train, test or stop, with dependencies and assumptions.
- Leave with the next step.A 90-day roadmap and one pilot or improvement brief: owner to confirm, baseline, success criteria and decision gates. A 60-minute findings call agrees the actions.
*Introductory review fee. Any applicable VAT, scope and dates are confirmed before you commit. Findings are targeted within seven working days after the sessions and receipt of agreed inputs. Implementation, licences, formal training and specialist assurance are scoped separately.
Implementation after the review
One workflow.
A result you can assess.
If a pilot is justified, we agree a separate scope, an operational owner and the evidence needed to continue. We can help with integration, team adoption and ongoing improvement.
You can also take the findings to your own team or existing supplier. Improving an existing feature or stopping an unsuitable pilot may be the right decision.
Business experience behind the advice
The people and process
matter as much as the tool.
Justin founded and led Seriun and built its marketing function from the ground up. His software delivery experience includes leading an office-supplies ERP programme for around 30 users, with a team of up to 10.
That experience informs how Fortuna Pace approaches implementation: understand the work, involve the people doing it and make responsibilities clear.
Meet JustinBefore we talk.
We have not started using AI yet. Is this useful?
Yes. We use the same workflow-first approach to compare opportunities, check the prerequisites and choose a manageable first pilot. You do not need to buy a tool before we talk.
We already pay for AI tools. Will you review them?
Yes. The scope includes up to three tools or pilots, looking at the work they support, quality, adoption, costs and practical implementation gaps.
Can you help us assess local models?
Yes. We consider the workflow, available hardware, quality requirements, information flows and who would run the service. The review can recommend local, cloud or hybrid use. Hands-on model testing, installation and deployment are separately scoped.
Will this work in our sector?
The review is for SMEs across all sectors. The first conversation checks the problem and scope. Any need for specialist expertise is identified before an engagement.
Does £1,495 include building or fixing the integration?
The fee covers the review, findings, roadmap and one pilot or improvement brief. Production integration, configuration, rollout and support are separately scoped.
Can you guarantee a saving?
No. We establish a baseline and assess the complete task, including checking and exceptions. The next step is chosen using evidence and clear assumptions.
Start with one real problem
Where is AI
adding friction?
Tell Justin what your team is trying to do and where the work gets stuck. If you are just starting, tell us which recurring task you want to improve.
We’ll reply within one working day to arrange a free 20-minute conversation. No technical brief needed.