How to Use Claude for B2B SaaS Accounting [Free Masterclass]

What You'll See

• How to build a SaaS revenue forecast in Excel without writing a single formula. Change the price, the growth rate, or the churn and watch the full model update in seconds.
• What happens when AI reads your contracts for you. See how a Claude skill extracts revenue recognition data from PDF contracts, applies ASC 606 logic, and outputs a clean table. The work that takes your team a full day, done in minutes.
• How to go from a messy folder of invoices to a completed bank reconciliation. Watch Claude Co-work organize invoices by month and currency, match them against a bank statement, and flag the exceptions.
• How to forecast financial statements from a SaaS pricing model. Start with your subscription revenue, layer in headcount costs and cash flow, and generate a 24-month outlook — all using Claude inside Excel.

Meet Your Host

Nicolas Boucher (Founder @ AI Finance Club)

The #1 AI in Finance influencer. 15 years in corporate finance at PwC and Thales. Has trained 10,000+ finance professionals at companies including KPMG, Mercedes-Benz, Siemens, and Carlsberg. His mission: help finance pros thrive in the era of AI.

Background:

• 15 years in corporate finance at PwC and Thales
• Trained 10,000+ finance professionals globally
• Worked with teams at KPMG, Mercedes-Benz, Siemens, Carlsberg
• #1 AI in Finance influencer

Who Should Attend

Controllers, revenue accountants, and senior accountants at B2B SaaS companies who are:

• Trying to close month-end books faster
• Working with complex contracts (usage-based pricing, hybrid models, multi-element arrangements)
• Currently using Excel for rev rec, reconciliation, or forecasting
• Curious about using Claude for day-to-day accounting tasks

About This Session

You didn't study accounting to copy-paste data between spreadsheets.

You did it to understand the business. To spot what others miss. To walk into a meeting and say "here's what the numbers are telling us" — and have people listen.

But right now, the close takes everything you've got. By the time you're done wrestling with rev rec, reconciling invoices, and writing variance commentary nobody reads, there's no time left for the work that gets you promoted.

And it's getting harder. Your sales team keeps closing deals with custom pricing. Every new contract is a puzzle. ASC 606 turned judgment calls into compliance risk. And the spreadsheets just keep multiplying.

This masterclass shows you how to use Claude to solve this. You'll see live Claude use cases that handle the mechanical parts of your job, so you can spend your time on the parts that require a human brain.

This isn't about replacing you — it's about freeing you up to do the work you were hired to do.

Whether you want to close faster, forecast with more confidence, or just have time to think before the next board meeting — this is the most practical hour you'll spend on Claude this month.

Webinary Summary

Finance teams are under constant pressure to close books faster, recognize revenue accurately, and do more with less headcount. In this webinar, Nicolas Boucher – corporate finance veteran and AI trainer for teams at KPMG, Mercedes-Benz, and Siemens – walked through five practical Claude workflows that take manual work off the finance team's plate today.

Here's a breakdown of every use case covered.

1. Invoice organization with Claude Cowork

Nicolas opened with a use case that resonated immediately: hundreds of PDF invoices sitting in a single folder, unorganized by date or currency. Using Claude Cowork – Claude's desktop app that can access your local files – he gave Claude permission to read the folder and reorganize every invoice into subfolders by month and currency, without opening a single file manually.

The entire restructuring happened in minutes. The key principle: Claude Cowork writes a mini-script that reads each PDF, extracts the relevant data, and moves the file accordingly. The one rule Nicolas emphasized – always keep a backup before giving Claude access to any folder, since it can rename or move files permanently.

2. Bank reconciliation

With invoices organized, Nicolas moved straight into reconciliation – uploading a bank statement and asking Claude to match it against all the invoices in the folder. Claude used a built-in finance skill (a reconciliation-specific prompt layer installed via the finance plugin in Cowork) to structure the output into a color-coded Excel file with matched, unmatched, and flagged items clearly separated.

The result was what Nicolas called the clearest reconciliation documentation he had seen from any source – human or AI. He noted that the finance plugin is free to install and adds pre-built skills for reconciliation, variance analysis, and general entry workflows.

For teams dealing with the downstream accounting side of reconciliation, Zenskar's revenue recognition feature automates journal entries and keeps deferred and recognized revenue schedules audit-ready without manual intervention.

3. SaaS revenue forecast in Excel

Nicolas demonstrated how to build a complete SaaS revenue forecast model in Excel using Claude for Excel (available on Pro, Max, Team, and Enterprise Claude plans via Add-ins). Starting from a plain-English prompt describing two pricing plans – a standard plan and a premium plan, each with per-seat components – Claude built a 12-month model with assumption tabs, formulas, growth/churn inputs, and an MRR graph, all in under six minutes.

He then showed how to ask Claude to audit its own work – generating a separate audit tab that documents every formula, flags potential errors, and explains the logic behind each KPI calculation. His advice: use graphs as a fast sanity check, ask Claude to simplify formulas if you want them easier to edit, and always treat AI output the same way you'd treat work from a junior team member – review it, don't just trust it.

4. Full financial model: P&L, balance sheet, and cash flow

Building on the revenue forecast, Nicolas extended the model into a complete financial model – headcount and cost tab, P&L, balance sheet, and cash flow statement – all linked and formula-driven, built in roughly six minutes. The prompt was more specific this time: synthetic data for now, average salary assumptions, fixed costs, and marketing as a percentage of revenue.

He showed how changing a single headcount input cascades through the entire model automatically, and re-used the self-audit technique to validate every tab before treating the output as final.

5. ASC 606 revenue recognition documentation using a custom skill

The final use case was the most technically relevant for SaaS finance teams. Nicolas walked through how he built a custom ASC 606 skill in Claude Cowork by:

• Asking Claude to research Big Four guidance on SaaS revenue recognition from the web
• Prompting it to synthesize that research into a structured skill (essentially a detailed system prompt)
• Using the skill to analyze three SaaS contracts and produce a fully documented revenue recognition workbook – performance obligations, transaction price allocation, recognition schedules, and all required ASC 606 checkpoints – for each contract

The output followed the five-step ASC 606 model automatically because the skill encoded that structure. Nicolas was clear about the limitation: this approach works well for a manageable number of contracts. At scale, you need a purpose-built platform.

That's exactly where Zenskar comes in. Unlike generic AI tools, Zenskar is built ground up on ASC 606 and IFRS 15 – automatically extracting performance obligations from contracts, allocating transaction prices, generating revenue schedules, and posting journal entries to your ERP without manual intervention at every step.

When Claude is enough – and when it isn't

Nicolas made an important distinction that Priyam echoed: general-purpose AI tools like Claude are powerful Swiss knives. They are excellent for automating manual tasks within existing workflows – organizing files, building models, drafting documentation. But when you're dealing with hundreds of contracts, real-time usage data, multi-entity setups, or audit-grade compliance requirements, you need a purpose-built system underneath.

Zenskar handles everything Claude demonstrated on the revenue recognition side – but at scale, automatically, and with full ERP sync. Finance teams don't even need to log into the platform; the same workflows can be triggered from Slack, Microsoft Teams, or directly from Claude via Zenskar's MCP integration.

If you're still managing revenue recognition in spreadsheets, this buyer's guide breaks down exactly when spreadsheets stop working and what to look for in a purpose-built alternative. And if you're evaluating order-to-cash platforms more broadly, the O2C software guide covers what to look for across the full billing-to-recognition stack.