Finance role

AI Financial Analyst

An AI financial analyst that turns your exports into models with the assumptions on top, three scenarios and a sensitivity table, and tells you what the numbers mean.

Give it your actuals, your budget and the decision in front of you, and it does the analysis a finance team would: variance reports that explain why you are off plan, cash-flow and runway forecasts with base, upside and downside cases, unit economics by customer or product, and investment cases where every assumption is written down and tested. It separates history from forecast, states its confidence, and says when a recommendation flips on a small change.

Things you could ask for

Written the way you would actually say them. The agent plans the steps itself.

  • “Here is last month’s P&L export and the budget in Google Sheets. Write the variance report: the five biggest variances, the likely cause of each, and what they change in the full-year outlook.”
  • “Build a 12-month cash-flow forecast from our bank export and the Stripe revenue data, with base, upside and downside cases, and tell me the month we would run out of cash in each.”
  • “We are deciding between hiring two salespeople or spending the same on ads. Model both with stated assumptions, show the payback period for each and the assumption that would change the answer.”
  • “Work out unit economics by plan from our Stripe subscriptions and the cost sheet: contribution margin, CAC payback and LTV, with the formulas left in the sheet.”
  • “Turn this quarter’s numbers into a five-slide board update in Google Slides: where we are against plan, why, the cash position and the decisions we need from the board.”

What you get back

Analysis you can check line by line: Google Sheets models with inputs, calculations and outputs on separate tabs and the formulas kept, variance reports with a root cause per line, scenario and sensitivity tables, and board-ready summaries in Google Docs or Slides that lead with the decision.

Where it is the wrong tool

  • It works from the data you give it. QuickBooks can be connected through the Connectors page; for Xero, NetSuite and other ledgers, export the trial balance or reports and share them.
  • It is not your accountant or auditor. It does not file returns, sign off accounts or give regulated investment advice; it prepares the analysis a qualified person decides on.
  • Forecasts are labelled as forecasts. Where an input is missing it states the assumption it used instead of hiding it in a cell.
  • It does not move money. Payments, payroll and transfers stay with you and your bank.

Starting an agent with this role

  1. 1Create a Burrak account — any plan works, and your first week is $1.
  2. 2In the Marketplace, open the Roles tab, find Financial Analyst and choose “Start an agent as Financial Analyst”. That creates a remote agent with the role’s instructions in its system prompt.
  3. 3Connect the tools it needs — Google Sheets, Stripe, Google Docs, Google Slides, Airtable, Google Drive — from Connectors.
  4. 4Give it a brief. It keeps working in the cloud after you close the tab, and you are only charged while it is actually working.

The Financial Analyst role, in practice

Will its spreadsheets be usable by someone else?
That is the point of how it builds them. Assumptions sit on one tab with their source, calculations reference them rather than hard-coded numbers, and each version is saved instead of overwritten, so your accountant or a board member can follow the logic and change an input.
Can it run the monthly variance report on its own?
Yes. Set it up as a scheduled task: once your export lands in Google Drive each month, it compares actuals to budget, writes the report and sends it to you, with the same structure every month so the trend is easy to read.
How is this different from the Pipeline Analyst?
The Pipeline Analyst looks forward at sales deals and forecast risk. The Financial Analyst works across the whole business — revenue, costs, cash and investment decisions — and is the one to ask whether the plan is affordable.

Capability reference

Summarised from the role’s instructions, which are adapted from the agency-agents collection (AgentLand Contributors), used under the MIT licence.

  • Three-statement models, DCF valuation and scenario analysis
  • Budget-versus-actual variance analysis with root-cause decomposition
  • Cash-flow, working-capital and runway forecasting
  • Unit economics: CAC, LTV, payback and contribution margin
  • Board and executive summaries that lead with the decision
Your first 7 days for $1

Your AI assistant is waiting.
Put it to work today.

Put an agent on the work that never needed a human in the first place — the research, the reports, the follow-ups — and take the week back.

$1 for your first week · Cancel anytime · Works on Mac, Windows, iPhone, Android & Web