Claude and Copilot training for finance professionals and teams.

You leave with the work done, and with the instructions and checks to do it again yourself.

What you can build

• A personal daily brief from your emails, diary items and the replies you owe
• A report ready for the meeting, from Excel to a dashboard or board pack
• Research you can trace, with sources and stated gaps
• Content that shows what you offer
• A website or a useful little tool
• A method you do not have to retype: a reusable skill or command

Starting points for a session, not a promise to cover everything in one sitting.

• Finding your way around Claude and choosing a suitable model and settings for the task
• Moving between chat, Cowork and Claude Code as the work requires
• Setting up projects, useful instructions and prompts for your own work
• Connecting approved email and files, including OneDrive or Google Drive where available
• Working with Excel, Word and presentations, using supported integrations
• Using connectors, plugins, skills and slash commands without giving away unnecessary access
• Setting up scheduled tasks, reusable design styles and a practical review routine

Sessions

What AI is · Using AI safely · Using AI in practice · AI in finance · Next steps

You leave able to

• Explain the difference between AI, machine learning and deep learning in plain words
• Tell generative AI apart from predictive AI, pick the right one for a job, and explain why its answers can be wrong
• Tell an AI agent apart from a chatbot or a copilot
• Know what may and may not go into a free tool, remove client details first, and know what prompt injection is
• Spot scams made with AI and aimed at staff, such as deepfake voice notes, fake invoices and phishing, and know where AI helps catch fraud
• Keep a human review step, own the final answer and spot where skewed data can bias it
• Write a complete prompt: role, context, data, the format you want and an example
• Ask AI to explain a variance, a ratio or a trend in your own numbers
• Check an AI answer against its source and your own numbers, and note who checked it
• Know what AI is already in Copilot, Sage or Xero, and use it to write a formula or clean a column in Excel
• Pick one task in your own work to try AI on first
• Know which tasks go, which roles appear, and how to stay useful

Sessions

Working with AI · Analysis and reporting · Risk and control · Making it work in your team

You leave able to

• Build reusable prompt templates with worked examples for recurring finance tasks
• Clean and check data before using AI, and practise on invented data, not client data
• Give the AI your own policies, IFRS notes and SARS guidance to work from, and check the passage it cites
• Use AI to prepare reconciliations and flag the items that do not match
• Draft a budget, build cash flow scenarios and run sensitivity tests with AI
• Tie totals to the trial balance, demand sources, catch invented numbers
• Turn a variance table into commentary and charts management can read at a glance, then into emails and slides
• Flag duplicate payments, odd journals and changed bank details
• Draft a risk and control matrix for your AI use, test the controls and keep the evidence
• Write the rules on tools, data, review and logging
• Put a use case, its risks, controls and return on one page for the MD or board, and test it against a colleague's challenge

• The applications the team uses, for example Sage, Xero, Excel or your ERP
• The problem areas, or simply that the team needs training and upskilling
• Whether the team works in Claude or Copilot

The course content stays the same. The examples, the exercises and the ways of working are built around your systems and your month.