FAQ

We’ll discuss your goal and the tools you already use, compare possible approaches and agree a practical next step with you. The initial review is complimentary. Detailed technical investigation and solution design are paid work, with the scope and fee agreed before they begin.

Software that takes over steps people repeat: reading an email, filling in a form, preparing a draft. AI handles the parts that need reading or judgement; the rest is ordinary automation.

Automation takes over repeated steps in the tools you already have. A custom AI system is built around a task that needs more than those tools provide, such as an assistant using your company knowledge or an agent working across systems. Sometimes we combine both in one solution. We’ll help you compare the options against your goal.

There are three parts: building the system, supporting it, and the services it runs on, such as AI models and subscriptions. The plan shows all three before work starts.

It can be worthwhile when the time saved or improvement in service justifies the cost. We agree on a useful starting scope and how to measure the result, then review it with you before deciding what comes next.

The timeline depends on what the system needs to do, the tools it connects to and the testing required. We agree the schedule with you before work starts.

We review the task with you and compare the options before choosing an approach and selecting the tools that best fit your goals.

Usually not. We connect to the CRM, inbox and spreadsheets you already use. If a tool can't be connected, we explain why and suggest options.

Common CRMs, email, spreadsheets and accounting software. We use n8n, other platforms or custom integrations depending on the task.

Mistakes are possible, so we plan for them. We test the system on your own examples, send uncertain cases to a person and keep a log of its actions, so anything can be checked and corrected.

Your team can manage it with clear instructions, or we can maintain it through monthly support. We’ll agree the support arrangement with you.

Yes. We agree a useful starting scope around your goal, then measure the result before deciding what comes next.

A chatbot answers. An agent also acts: it looks up data, updates records and prepares documents within your rules.

Retrieval-augmented generation: the AI first finds the relevant passages in your knowledge base and answers from them, with the source attached.

We choose for each task and often combine several in one system.

It depends on the scope, the integrations and how much the AI models are used. Before work starts, you get a projected breakdown of build, support and running costs.

Each system gets access only to what the task needs. We agree where your data is stored and which AI services can process it.

Yes, if it's feasible. We first check whether an existing tool can do it; if not, we design and build the system for it.