Practical Microsoft 365 Copilot training for finance teams

AI for finance with Microsoft 365 Copilot

Apply one transferable prompt method across financial research, data cleaning, reconciliation, cash-flow scenarios, management reporting and meeting follow-up. Copilot supports preparation and analysis; authorised finance professionals retain control of calculations, postings, credit, reporting and approval.

Request a proposal for your finance team View the programme

One interactive daySix governed finance tasksReconcile, review and approve

Governed finance workflow

From approved source data to reviewed reporting

Approved source dataCurrent, authorised and appropriately classified finance information

Prompt methodRole, Context, Objective and Complexity

Analysis or calculationDraft cleaning steps, reconciliation logic, scenarios or summary

Reconciliation and controlsCheck totals, formulas, units, mappings, assumptions and exceptions

Human validationFour-eyes review and authorised finance judgement

Reporting or actionControlled communication, not autonomous posting or approval

Finance remains accountable

Copilot can prepare drafts, analyses and scenario structures. It must not independently approve postings, payments, provisions, credit decisions, forecasts, closes, audit conclusions or financial reports.

TraceabilityLink conclusions to sources, calculations and assumptions

ControlsReconcile formulas, classifications, scenarios and totals

AuthorityNo autonomous posting, payment, credit or report approval

In short

What to know before choosing this workshop

A practical one-day finance training covering credit-risk research, reconciliation, cash-flow scenarios, management summaries, reporting and controlled meeting follow-up.

For: CFOs, financial controllers, accountants and FP&A analysts want practical workflows. · Finance business partners, reporting, consolidation, treasury or operations teams need clearer preparation and review.
Format: Approximately seven hours, adapted to your team
Languages: English, French or Dutch
Main tools: Excel, Microsoft 365 Copilot Chat, Word, PowerPoint, Outlook, Teams, Microsoft 365 Copilot Notebooks, Agent Builder
What participants will be able to do: Structure and clean raw financial datasets more systematically. · Analyse reconciliation differences and open items more clearly.
Access: Features and licences depend on your Microsoft 365 environment
Human review: Participants verify outputs; accountable decisions remain with people

WorkWithAI Belgium

A clearly scoped Copilot workshop

WorkWithAI is a Belgian provider of practical Microsoft 365 Copilot training, delivered by Arnaud Llamas Bravo in English, French and Dutch.

Finance analysis, reporting and decision support with explicit verification and control boundaries.

Designed for: Finance professionals, controllers, analysts and finance leaders.
Business tasks: Analyse and reconcile financial data
Explain variances and scenarios
Prepare management and board reporting
Choose this workshop when: financial accuracy, traceability and review matter more than general office productivity.

WorkWithAI

Sources and product notes

Primary sources supporting product, privacy and governance statements on this page.

How Microsoft 365 Copilot works ↗Microsoft Learn · 2026-07-22
Security for Microsoft 365 Copilot ↗Microsoft Learn · 2026-07-22
Central Balance Sheet Office ↗National Bank of Belgium · 2026-07-22

Recognition

Finance work is spread across data, files, meetings, controls and reporting cycles.

The opportunity is not autonomous accounting or audit. It is a controlled workflow that makes data quality, assumptions, reconciliations, evidence and approvals easier to inspect.

Too much time is spent on data cleaning, reconciliation, reporting and forecasting.

Financial information is spread across Excel, Word, Outlook, Teams and other sources.

Moving from raw data to management insight is difficult and repetitive.

Reports and dashboards require repeated manual updating and explanation.

Conclusions, calculations and scenario assumptions are not always traceable.

Board packs and CFO summaries are prepared manually under time pressure.

Too little time remains for analysis, advice, challenge and strategy.

Teams remain uncertain about confidentiality, controls, compliance and responsible AI use.

Desired outcome

Prepare finance analysis more systematically without delegating finance authority.

Results depend on source quality, data models, permissions, definitions, organisational policy, review controls and professional judgement.

Structure and clean raw financial datasets more systematically.

Analyse reconciliation differences and open items more clearly.

Build base, downside and upside cash-flow scenarios with explicit assumptions.

Structure credit-risk information using traceable public sources.

Prepare reviewed management and CFO summaries.

Visualise KPIs, variances, runway and scenarios more clearly.

Summarise finance meetings with decisions, actions and owners.

Centralise a bounded finance dossier in Microsoft 365 Copilot Notebooks.

Prototype a simple reusable finance assistant with explicit authority limits.

Core finance tasks

Six practical workflows with explicit control and authority boundaries.

Each task shows its purpose, required source data, Copilot support, concrete output, human validation and the main control risk.

Credit-risk report

Structure a traceable view of a company and its financial context.

Source-linked working credit-risk report and open-question list.

Data cleaning

Make a raw financial dataset more usable for controlled analysis.

Documented cleaning plan, exception log and reviewed transformed dataset.

Reconciliation report

Explain differences between bank, ledger or subsystem balances.

Reconciliation working paper with evidence links, owners and unresolved items.

Cash-flow forecast

Prepare liquidity scenarios using explicit assumptions.

Three-scenario cash-flow model and assumption register.

CFO summary

Translate reviewed finance analysis into a concise management narrative.

Reviewed CFO summary, chart outline and management-slide structure.

Finance meeting notes

Capture decisions, actions, owners and unresolved questions.

Reviewed minutes, decision log and action list.

Copilot across the finance cycle

Use the application that matches the finance task.

Availability, licences, service plans, rollout, permissions and administrative controls must be confirmed for the organisation.

Excel

Data cleaning, bank reconciliation, cash-flow forecasts, ratios, variances, sensitivities and scenarios.

Verify formulas, ranges, units, signs, classifications, assumptions and control totals; no autonomous posting or approval.

Microsoft 365 Copilot Chat

Financial research, credit-risk preparation, sector context and multi-source synthesis.

Verify source authority, currency, dates and citations; work and web grounding depend on licence and configuration.

Word

Credit-risk reports, reconciliation notes, finance policies, summaries and translations.

Word does not replace accounting, audit, legal, tax or compliance review.

PowerPoint

CFO summaries, board packs, P&L, cash-flow, balance-sheet and KPI dashboards.

Review every figure, chart, label, comparison and narrative; board-ready does not mean approved.

Outlook

Communication with management, auditors, stakeholders and process owners.

Review recipients, attachments, confidentiality, tone, commitments and formal approval before sending.

Teams

Finance, audit, budget and close-meeting recaps, decisions and actions.

Review transcript access, attribution, confidentiality, decisions, retention and action ownership.

Microsoft 365 Copilot Notebooks

Centralise selected definitions, methodologies, source files and dossier references.

Maintain need-to-know access, ownership, source currency, scope, versioning and retention.

Agent Builder

Prototype bounded data-cleaning, credit-risk or reconciliation assistants.

No posting, payment, provision, credit, forecast or report-approval authority; testing and lifecycle governance remain required.

Learning outcomes

Practical finance capabilities within a one-day training.

The course supports preparation, analysis and control habits; it does not promise finance mastery, audit-ready status, guaranteed accuracy or certification.

Write prompts using Role, Context, Objective and Complexity.

Select finance sources and make outputs traceable.

Structure a credit-risk report using public, verifiable sources.

Clean raw financial datasets with a documented exception log.

Analyse and report reconciliation differences and open items.

Build a three-scenario cash-flow forecast with sensitivity and runway.

Prepare a CFO summary with P&L variances, cash flow, balance sheet and KPIs.

Rewrite and translate finance text while preserving definitions and figures.

Summarise finance meetings with reviewed decisions and actions.

Centralise a bounded finance dossier in Microsoft 365 Copilot Notebooks.

Prototype a simple finance assistant with Agent Builder.

Apply confidentiality, traceability, controls, four-eyes review and human approval.

Programme of the day

From traceable finance research to a bounded finance assistant.

09:00Welcome and kick-off: Microsoft 365 Copilot and Golden Rules

09:45Prompt method: Role, Context, Objective and Complexity

10:30Microsoft 365 Copilot Chat and Word: credit-risk report using traceable sources

11:15Excel: data cleaning

12:00Lunch break

13:00Excel: bank reconciliation and cash-flow forecast with scenarios

14:00PowerPoint: operational CFO summary and KPI dashboards

14:45Outlook and Teams: finance communication and meeting recap

15:30Microsoft 365 Copilot Notebooks: centralise a finance dossier

16:00Agent Builder: build a bounded finance assistant

16:40Wrap-up, action plan and Q&A

Responsible use and governance

Traceability, reconciliation, controls and human finance judgement remain visible.

Copilot supports preparation. Authorised professionals remain responsible for data, calculations, accounting treatment, credit, audit, reporting and decisions.

This page does not provide financial, accounting, audit, legal, tax, credit or compliance advice. Final decisions remain with authorised professionals and governance bodies.

Data minimisation and least-privilege access.

Separation of demo, test and production data.

Approved finance sources, data classifications and secure storage only.

Source currency, ownership, permissions, period and version control.

Document formulas, mappings, assumptions, scenarios and definitions.

Reconcile calculations, totals, units, signs and source-to-output differences.

Make missing, contradictory or unresolved information explicit.

Human validation before reporting, posting, payment, credit or decision.

Four-eyes review for material or externally shared finance output.

Audit trail, logging, versioning and evidence retention where required.

Protect personal data, bank information, customer data and confidential figures.

Review bias and inappropriate proxies in credit or risk assessment.

Use IFRS, Basel III and internal policy references only in their correct scope with expert validation.

Escalate to finance, accounting, audit, legal, tax, risk, compliance, IT, privacy or security.

Test, own, monitor, update and retire reusable finance assistants.

Do not use real confidential financial data in public demos or screenshots.

Audience and suitability

For finance teams seeking a controlled Microsoft 365 Copilot workflow.

Suitability depends on role, data access, organisational policy, finance controls and the confirmed Microsoft environment.

Suitable when

CFOs, financial controllers, accountants and FP&A analysts want practical workflows.

Finance business partners, reporting, consolidation, treasury or operations teams need clearer preparation and review.

The organisation can provide fictional, anonymised or explicitly approved finance data.

Participants can work within existing finance, privacy, audit and security controls.

The goal is human-reviewed analysis, reporting and reusable assistance.

Less suitable when

A complete accounting, IFRS, Basel III, treasury or audit course is required.

The expectation is automated credit decisions, postings, payments or close approval.

The organisation expects error-free figures, guaranteed forecast accuracy or a faster close.

The need is Microsoft 365 administrator training or formal finance certification.

Professional accounting, audit, tax, legal, risk or compliance judgement is expected to be replaced.

Arnaud Llamas Bravo presenting Microsoft Copilot during a professional workshop

Arnaud Llamas BravoFounder & trainer, WorkWithAI

Photo 1 of 3

Your trainer

Arnaud Llamas Bravo

Clear.Practical.Human.

Since January 2024, I have trained more than 8,500 professionals. I am passionate about helping individuals and organisations build the skills they need to use AI effectively. My practical, hands-on approach ensures that every participant leaves with tools and ideas they can apply immediately.

I specialise in delivering trilingual training in English, French and Dutch across recruitment, human resources, education, marketing and sales. My mission is to make AI accessible, useful and transformative for teams that want to work smarter and achieve measurable results.

Through collaborations with renowned organisations including Proximus, Honda and the Flemish Government — and by creating tailored programmes for different sectors — I combine real-world experience with an in-depth understanding of AI’s potential.

Connect with Arnaud on LinkedIn Read all participant testimonials

8,500+ professionals trained247 verified participant storiesTraining in English, French and Dutch

Written and reviewed by Arnaud Llamas Bravo · Editorial methodology · Updated 27 July 2026

Practical information

Known training facts and items requiring confirmation.

Training: AI in Finance with Microsoft 365 Copilot
Duration: One full interactive day of approximately seven hours, including a lunch break
Audience: CFOs, financial controllers, accountants, FP&A analysts and finance teams
Applications: Excel, Word, PowerPoint, Outlook, Teams, Microsoft 365 Copilot Chat, Microsoft 365 Copilot Notebooks and Agent Builder
Modules: Credit risk, data cleaning, reconciliation, cash-flow scenarios, CFO summary and KPI dashboards
Prior knowledge: No technical prior knowledge required; finance context supports the exercises
Participation: A laptop and confirmed access to the required Microsoft 365 Copilot capabilities are recommended or required for hands-on work
Practice data: Use fictional, anonymised or explicitly approved finance data under organisational policy
Format: Interactive and hands-on; final delivery format requires confirmation
Governance: Finance controls, privacy, security, audit and professional review remain applicable

Frequently asked questions

Practical, licensing and governance questions.

Who is this finance training for?

It is intended for CFOs, controllers, accountants, FP&A analysts, finance business partners, reporting, consolidation, treasury and finance-operations professionals seeking practical Microsoft 365 Copilot workflows.

Is technical or finance prior knowledge required?

No technical prior knowledge is required. Familiarity with finance processes and the organisation’s data and control environment supports the exercises.

Which Microsoft 365 Copilot licence is needed?

Exact licences, service plans, tenant configuration and feature availability must be confirmed before delivery. Access is not identical for every user or organisation.

Can I attend without active Copilot access?

Participation may still be possible in an observation or shared-demonstration format, but hands-on exercises depend on the confirmed environment and delivery setup.

Can I work with my own finance data?

Only when organisational policy, permissions, confidentiality, privacy and security controls allow it. Fictional, anonymised or explicitly approved data is preferred.

How is financial information protected?

Use data minimisation, least privilege, approved storage, access control, secure sharing and the organisation’s existing finance, privacy and security policies. No absolute security claim is made.

Can Copilot approve postings, payments or credit decisions?

No. Copilot and reusable assistants have no independent authority to post, pay, approve credit, set provisions, sign off forecasts or approve reports.

How reliable are forecasts, reconciliations and credit reports?

They are working outputs that depend on source quality, formulas, assumptions, definitions and review. Reperform calculations, reconcile totals and escalate missing or contradictory information.

Are reports automatically IFRS- or Basel-compliant?

No. IFRS or Basel references require correct scope, current approved wording and review by qualified finance, accounting, risk, audit, legal or compliance professionals.

Which public sources may be used?

The National Bank of Belgium Central Balance Sheet Office provides public annual-account information. Third-party services such as jaarrekening.be and Companyweb may also exist, but access, licences, terms, coverage and currency must be checked before use.

What is the role of human validation?

Authorised professionals verify sources, formulas, assumptions, scenarios, reconciliations, accounting treatment, materiality, narrative and final decisions.

What is the difference between Copilot Chat, Notebooks and Agent Builder?

Copilot Chat supports conversational research and synthesis; Copilot Notebooks gather selected references for a bounded workspace; Agent Builder creates reusable declarative assistance. Availability and governance differ by licence and configuration.

Will I build a finance assistant?

The programme includes prototyping a bounded assistant for a recurring finance task. It remains subject to testing, access control, ownership, monitoring and explicit authority limits.

Can the training be delivered in-company or adapted?

That delivery option may be discussed, subject to approved scope, environment, data policy and commercial arrangements.

In which languages is the training available?

English, French and Dutch versions are represented on this page. Actual delivery languages must be confirmed.

Which materials or follow-up are included?

Materials, support and follow-up must be confirmed in the approved proposal. No unconfirmed items are promised.

What participants say

Practical learning you can apply immediately at work.

Participants value the clarity, energy and immediate application of every WorkWithAI workshop.

“My name is Julien Vansantvoort, and I am Director of Mewa Belgium. In my role, I am closely involved in financial management, operational performance, and strategic decision-making, so understanding how AI can support finance functions is highly relevant to my day-to-day responsibilities. I found the workshop extremely enriching, relevant, and inspiring. The content was concrete, insightful, and directly connected to …”

Julien VansantvoortDirector Mewa Servibel

“The training on artificial intelligence and its applications in finance largely met my expectations. As a CFO, I primarily hoped to understand how AI can simplify repetitive tasks and create more room for analysis and well-founded decision-making. The course had strong content, was clearly structured and was delivered in a particularly engaging and accessible way. In addition to a good …”

Ivan VervoortCFO Group Joos

“I already had an idea of what AI could mean for the financial world, but during this workshop, I was still left amazed more than once. It was impressive to see what is already possible today. At the same time, it left me with mixed feelings: on the one hand, the enormous added value AI can bring, and on the …”

Sofie SmetsCFO Vlinvesta

“The ‘AI in Finance’ course led by Arnaud Llamas is an absolute recommendation for every finance professional who takes the future of the field seriously. At a time when artificial intelligence is no longer a trend but a strategic necessity, this training offers exactly the right mix of in-depth content, practical applicability and strategic insight. Arnaud has a unique ability …”

Klaartje CeyssensCFO - Marmo

“As an accountant in an SME, I had already heard a lot about AI, but this workshop genuinely opened up a completely new world for me. I honestly did not know that you could do so much with it or that it could be so practical in everyday working life. After the training, I immediately started using the prompts we …”

EvelienBoekhoudster bij Symo

“I recently had the opportunity to complete a training program on recruitment and ChatGPT, and I must say it was an incredibly enlightening experience. The training was not only good; it was outstanding. The depth of knowledge and practical insights I gained during this training were invaluable. I learned how to leverage ChatGPT's capabilities in the recruitment process, from crafting …”

Barbara HendrickxSplendit

Go to testimonial

50 selected testimonialsSelected from 247 approved participant stories across practical AI and Copilot workshopsView the public source ↗

Tell me what your team needs.

Share the essentials and I’ll get back to you with the most useful next step.

Name *Work email *Organisation *Phone (optional)Workshop or topic *Business challenge or additional context (optional)

I agree that WorkWithAI may use these details to answer my request. I have not included confidential or sensitive information. Privacy policy.

Written and reviewed by Arnaud Llamas Bravo · Last reviewed 22 July 2026