Overview
Four cost centre heads ask the same question every month: where does our spend sit against target? The answer sits in Power BI, and someone in FP&A opens the report and types out the numbers by hand. The monthly pack is rebuilt from nothing every cycle.
Two days of building on Gemini Enterprise. Day 1 starts in Copilot Chat, sharpening how you brief it, before the day moves on to Gemini Enterprise for your first agent build. Day 2 grounds an agent on a published extract of your own figures, and builds a second that drafts the monthly pack on schedule and holds for your approval before it sends. You leave with both.
Who should attend
- FP&A, forecasting and budgeting staff who read Power BI reports every day
- Finance managers who sign off what those teams produce
- Participants with no prior agent-building experience
What you will be able to do
- Brief Copilot Chat with RCR and 5W1H, then use uploads, canvas and reverse prompting to get more from it.
- Tell Gemini Enterprise apart from similarly named Google products, and judge what a connector lets it see.
- Decide whether a task deserves an agent, then build and share a grounded agent in Agent Designer.
- Publish a Q&A agent grounded on your budget snapshot.
- Build a scheduled agent that drafts the monthly pack and waits for your approval.
- Place your recurring work on an Effort and Frequency grid and rebuild one FP&A task as an agent.
Course content
Day 1: Sharper prompting, Gemini Enterprise and your first agent
- RCR, 5W1H and advanced chat functions in Copilot Chat: Role, Context and Request built into a full brief, expanded with the 5W1H checklist, then uploads, the canvas Page and reverse prompting.
- Gemini Enterprise search and chat: which Gemini product this is, permission-aware search across connected Microsoft 365 data and what a connector does and does not let it see, and choosing between the fast model and the deep one.
- Your first agent in Agent Designer: the two-question test that decides whether a task deserves an agent, writing the grounding rule into the Instructions, and sharing the agent with a colleague to find what breaks.
Day 2: Connected agents and the rebuild
- Advanced agents, connectors, actions and schedules: a spending Q&A agent grounded on a published budget-versus-actual extract, where Power BI fits (agents read a published extract, never the live report), and a second agent that drafts the monthly pack on schedule and waits for a named owner's approval before it sends.
- Capstone, rebuilding one FP&A workflow: place your recurring work on the Effort and Frequency grid, put the top task through the two-question test, build the agent with the trainer beside you, and show the room what it saves.
How it runs
- Hands-on building with interactive lectures, live demonstrations and one running case study.
- Bring a laptop signed in to your work account, with Copilot Chat access, Gemini Enterprise access and agent creation enabled by your administrator.
- No coding is needed.
- The capstone is presented to the room, with peer review. Delivered in person.
