Sigma for finance teams
Close, explain, and reforecast in a spreadsheet your team already knows, on the numbers your data team already trusts.
Explain the variance
Variance commentary sits next to the numbers it explains. An agent can draft it for the lines that moved, and owners review and sign off.
Reforecast with owners in the loop
Business owners enter assumptions in the report, and they’re saved to the warehouse instead of emailed back.
Works like the spreadsheet you know
Formulas your team already writes on the full ledger, with Sigma Assistant to draft the rest when you ask.
Variances explained
The books close before the explanations do
Finance spends the days after close downloading reports, chasing commentary by email, and rebuilding the forecast in a file nobody else can audit.
| Today | With Sigma |
|---|---|
| Monday mornings spent downloading and stitching reports | Actuals arrive live from the warehouse with nothing to download |
| Variance commentary collected by email and pasted into decks | Variance commentary lives next to the numbers it explains |
| Assumptions buried in personal spreadsheet tabs | Owners save assumptions and commentary to the warehouse from the app |
| Forecast models that slow to a crawl as the file grows | Spreadsheet models that run on the full ledger |
| Forecast versions multiply with no record of who changed what | Each forecast submission recorded with who approved it |
| Business owners wait on finance for every number | Owners explore their own numbers on the same data finance uses |
Finance work on the numbers your data team trusts
Your warehouse already holds the ledger. Finance can work on it directly in a spreadsheet without an extract.
Your data team already built it
The warehouse already holds the ledger, the sub-ledgers, and the operational data. Finance works on it directly, and nobody makes an extract.
Start with the report everyone dreads
Build the one report that eats the first week of close. There’s no migration to finish first, and the next workflow can wait until this one works.
Formulas your team already writes
It works the way a spreadsheet does. Sigma Assistant drafts a formula, a pivot, or a whole workbook when you describe it.
Explain it, decide, submit
Explain the variance, decide whether it changes the outlook, and submit the revised forecast, all on the same live data.
Month-end review
Review results and collect explanations
Review financial results as soon as the books close. Drill from the P&L to the transactions behind a number, and collect explanations from business owners in the same app.
- Budget vs. actual by entity, department, and account
- Exceptions flagged for owner review
- Owner commentary saved to the warehouse
Scenario forecasts
Model scenarios and submit the forecast
Change a driver and watch revenue, margin, and cash update on live warehouse data. When the scenario holds up, submit the revised forecast for approval.
- Driver-based scenarios for revenue, margin, and spend
- Assumptions saved with who changed them
- Forecast submissions routed for approval
Headcount plans
Plan headcount against the budget
Hiring managers submit requests against the plan, see the cost against the budget as they type, and finance reviews them in the same app.
- Headcount modeled across growth scenarios
- Approval workflows for compensation budget
- Actual headcount tracked against plan
Spend control
Let an agent collect the expense forecast
An expense forecast agent collects parameters from budget owners, builds the forecast from operational drivers, and routes it for review.
- Driver-based models that categorize and predict spend
- Budget owners update monthly forecasts directly
- One live view of total spend for finance
Collections
Prioritize collections and track follow-up
Work aging invoices and unapplied payments on live data. Collectors assign accounts, add notes, and log promises to pay next to the invoice, without waiting on an ERP export.
- Live invoice aging by customer
- Assignments and notes saved to the warehouse
- Agents flag accounts sliding past terms
Assign Amberleigh Health to yourself.
AR aging · as of Oct 23
Collections
Past terms
$306.5K
+$48.2Kvs last week
Unassigned
$184.0K
1 account
Promised to pay
$41.9K
1 account
Accounts past terms
| Customer | 61–90 | 90+ | Owner | Status | Action |
|---|---|---|---|---|---|
| Amberleigh Health | $184,000 | $0 | Unassigned | Open | |
| Corlett Retail | $62,400 | $18,200 | J. Park | In review | |
| Corwin Logistics | $41,900 | $0 | J. Park | Promise to pay |
Collection activity
| Customer | Action | Amount | Due | |
|---|---|---|---|---|
| 1 | Corwin Logistics | Promise to pay | $41,900 | Oct 31 |
| 2 | Corlett Retail | Call logged | $80,600 | |
| 3 | Halden Foods | Paid in full | $22,300 | |
| + | ||||
61–90 days by customer
Agents with a specific job
Agents draft the commentary for owners to approve
Gathering context and drafting explanations are jobs you can check. Agents handle them, and nothing is final until an owner approves it.
- 1Gather contextWhen a line moves, an agent pulls the drivers from the GL, the plan, and the operational data behind it.
- 2Draft the explanationIt drafts commentary for the owner to edit, linked to the rows it used.
- 3Approve and recordThe owner approves, and the explanation is saved to the warehouse with who signed off.
Finance teams running on Sigma

75% employee adoption. 80% reduction in reporting time.How Yamaha Finance Runs Its Entire Financial Modeling Process in Sigma

Explore finance apps built in Sigma
Working examples from the App Library. Open one to see how it works.
Budget and Variance Approvals
Analyze budget vs. actuals on live data. Track variances, update forecasts, and route approvals in a governed AI app.
See the appExpense forecast agent
An agent collects parameters from budget owners, analyzes historical trends, and builds the forecast on live data, with the logic documented and an approval-ready memo.
See the appHeadcount & Compensation Planning
Plan and forecast headcount and compensation on live data. Model multi-year salary, bonus, and expense impact in a governed AI app.
See the app
Controls finance can sign off on
Built for teams that need auditability, access control, and one set of numbers that updates when the source changes.
Assumptions saved to the warehouse
Assumptions, commentary, and submissions entered in a report are saved to tables in your warehouse, with a full edit history.
Approval trails
Forecast submissions and approvals are recorded with who approved them and when.
Per-user OAuth
Connect with OAuth so queries run with each person’s own warehouse role.
Row- and column-level security
Reads and writes both inherit your table policies.
Audit trail
Audit logs record user activity, AI conversations, and edits to entered data, so you can see who changed what.
Your data stays put
Sigma queries your warehouse in place and never copies the data out.
Finance FAQs
How does Sigma change the month-end close?
Teams review results on live warehouse data as soon as the books close, investigate exceptions down to the transaction, and collect owner explanations in the same app instead of by email.
How long does it take to get started?
Most teams start with one report, usually the one that eats the first week of close, and build it on data already in the warehouse. There’s no migration to finish first, and you add the next workflow once the first one is working.
Do we need to learn something new?
If your team writes spreadsheet formulas, they can build in Sigma. The interface works like a spreadsheet, and Sigma Assistant can draft a formula, a pivot, or a whole workbook when you describe what you need.
Can we build collaborative budgets without emailing spreadsheets?
Yes. Budget owners enter their numbers and assumptions in the report, they’re saved to your warehouse, and finance sees submissions and edits in one place.
How do we track and explain budget variances?
Compare budget to actuals by any dimension on live data. Agents can draft explanations for the largest variances, and owners edit and approve them before they’re saved.
How does Sigma connect to our ERP?
Sigma reads ERP data once it lands in your cloud data warehouse, so finance works on the same tables as everyone else, with no separate extract.
Can each budget owner see only their own numbers?
Yes. Sigma inherits your warehouse’s row-level security, so each GM or budget owner sees and edits only their own cost center, in the same app finance uses.
How does Sigma handle what-if scenarios?
Change a driver in a scenario and the dependent calculations update. Scenarios save to the warehouse, so you can compare them and submit the one you choose.
Can treasury forecast cash on live data?
Yes. Model inflows and outflows on live receivables, payables, and inventory, and consolidate inputs from finance, sales, and operations in one forecast.
Can our models run on the full general ledger?
Yes. Sigma runs calculations as SQL in your warehouse, so models work on the full ledger instead of a file that slows down as it grows. Analysts keep a spreadsheet interface and the formulas they already know.
Bring next month’s variance review
We’ll walk through it on your warehouse: explanations drafted, owners in the loop, and the forecast submitted.





