Planning & Modeling
Financial planning and modeling: plan forward with a model you can trust.
Financial planning and modeling turns drivers, assumptions, and live actuals into one forward view of the business, so every decision gets tested before it gets made.
Actuals land, the forecast re-plans, and every path stays priced. Trace to source →
Start here
Financial planning and modeling, defined.
One model, every decision
Financial planning and modeling is the discipline of building a forward-looking model of a business: connecting operational drivers such as headcount, pipeline, and pricing to revenue, expenses, and cash, then using that model to budget, forecast, and test decisions before they are made. Planning sets the targets. Modeling supplies the logic that shows how the business behaves when assumptions change.
The strongest form is driver-based: change one assumption, ten sales hires or a pricing move, and revenue, spend, and cash recalculate together. Read the full guide to driver-based planning for how those models are built.
Operational drivers
Decisions, tested before they are made
A plan you cannot interrogate is a document. A model you can interrogate is a decision tool.
Why Cube
Plan faster. Defend every number.
Cube is the Agentic Finance Layer for financial planning and modeling: driver-based plans, rolling forecasts, and scenarios built on one governed model, inside the spreadsheets finance already uses, with FP&Agents that draft the forecast and show their work. One governed model under the plan changes what planning feels like. Three things finance teams get back.
Plan faster than the business moves.
Rolling forecasts, driver changes, and scenario comparisons in minutes, because actuals land in the model automatically and every assumption recalculates downstream.
Your team, out of the busywork.
The Planners draft the baseline forecast and build the scenarios. Your analysts spend the cycle on judgment calls, not on rebuilding spreadsheets.
Numbers you can defend.
Every figure in the plan traces to a source transaction or a named assumption. When the CEO asks where a number came from, you have the answer.
Plans rarely fail in the model. They fail in the handoffs.
The mechanism
How the financial planning loop works.
Every planning process runs the same loop: actuals and drivers flow in, the model applies your logic, and budgets, forecasts, and scenarios flow out. The difference between teams is how much of that loop is automated. This is the loop Cube's platform runs continuously.
Cube
One governed model for every plan, forecast, and scenario.
- One governed modelDrivers, logic, and actuals in one place
- Drivers you defineLogic written once, governed centrally
- Scenarios recalculated end to endChange a driver and every case updates
- Forecasts refreshed as actuals landNew actuals sync in, the plan re-baselines
In the spreadsheet
A plan that runs on live actuals.
Like the spreadsheet your team already runs, with one difference: actuals arrive live from the governed model, and plan versions publish back to it.
| FY27 $K | Q1 · Actuals | Q2 · Plan | Q3 · Plan |
|---|---|---|---|
| Sales reps (driver) | 24 | 28 | 30 |
| New bookings | 1,440 | 1,680 | 1,800 |
| Renewals | 2,210 | 2,340 | 2,470 |
| Total revenue | 3,650 | 4,020 | 4,270 |
| Operating expenses | 2,890 | 3,050 | 3,140 |
| Operating income | 760 | 970 | 1,130 |
In your browser
Planning, forecasting, and scenarios on one dashboard.
The planning dashboard reads from the same governed numbers as the grid above, so a driver change upstream moves every tile here.
Commentary · drafted for review
The Q3 base case holds at $4,270K revenue on 30 sales reps, with operating income stepping to $1,130K. The downside case at 26 reps gives up $240K of revenue; the upside at 34 reps adds $240K.
Revenue · Q3 plan
$4,270K
base case · 30 sales repsOp income · Q3 plan
$1,130K
up from $760K Q1 actualQ1 actual to Q3 plan walk
FY27 revenue $K · axis starts at $3,400K · green = planned growth
Opex mix · Q1 actual
- Payroll & benefits$1,940K
- Cloud & software$410K
- Marketing programs$360K
- T&E & other$180K
$K · sums to $2,890K · each slice opens to accounts
Revenue by quarter · scenarios
$K · Q3 dots = upside 34 reps / downside 26 reps · axis starts at $3,400K
Planning dashboard · FY27 · refreshed as actuals land · every tile opens to its transactions
Forecast view
The forecast as a trajectory: plan vs. actuals.
Quarterly revenue from the grid above, with the operating income line it produces. Each bar drills to the source transactions behind it.
One governed model, inside the spreadsheet your team already knows.
AI at Cube
Four agent teams. One superagent.
Charlie, Cube’s superagent, plans the work, orchestrates every team, and returns the decision-ready answer. For planning and modeling, the heavy lifting lands with the Planners →: forecasts, scenarios, and models built and stress-tested alongside you. 59% of CFOs expect to reduce finance headcount over the next three years (Battery Ventures, 2026); the planning work that stays human is judgment, and the agents take the assembly.
Plans the work, orchestrates every team, and returns the decision-ready answer.
Cube proposes.Finance approves.
Sales reps by Q3: 26
Sales reps by Q3: 30
Sales reps by Q3: 34
One driver changed: sales reps. Bookings, revenue, and operating income recalculate everywhere downstream.
Connect to Cube from Wherever You Work
Connect to Cube's single source of truth from where you work every day, without ever having to learn another tool.
Cube Workspace
One command center for planning, reporting, and analysis. No tab-hopping.
Explore the Workspace →| Account | Plan | Actual | Var |
|---|---|---|---|
| Marketing | $1.20M | $1.34M | +$142K |
| Events | $310K | $402K | +$92K |
Excel & Google Sheets
Your model stays yours. Cube keeps every cell current and traceable.
Explore spreadsheets →
PowerPoint & Google Slides
When the numbers change, the deck changes. No re-pastes before the board meeting.
Explore Cube Decks →AI assistants
Every insight maps back to the GL, so AI answers are safe to act on.
Explore AI via MCP →
Trust & controls
The plan, governed and secured.
The plan is one of the most sensitive documents in the company. Cube treats it that way, with controls your auditors can verify. Read the full security overview.
Role-based access
Plan owners see and edit only what they should, down to the account and department.
Read-only connections by default
Source systems stay the system of record. Cube reads; your ERP is never at risk.
Every change logged
Versions, edits, and approvals carry a who, a when, and a what, permanently.
Governed definitions
One set of metrics and hierarchies, so every plan and report agrees on what the words mean.
Trust in a plan is built the same way it is lost: one number at a time.
Trace to the transaction
Drill every planned number to its source transaction.
The $2,890K of Q1 operating expenses in the grid above drills to its accounts, and each account to the source transactions behind it.
| Feb 04 | AWS | usage, prod & staging | NetSuite | $86K |
| Feb 18 | Snowflake | annual true-up | NetSuite | $54K |
| Mar 09 | Datadog | renewal | NetSuite | $38K |
| + 41 more transactions · $232K | ||||
Proof
What finance teams say about planning in Cube.
“Cube has completely streamlined our budgeting, forecasting, and reporting. It’s fast, flexible, and integrates seamlessly with our systems, making it a true game-changer for our FP&A process.”
“Cube has enabled us to drill into budget and forecast-to-actual variances with a level of detail we simply couldn’t achieve before. We can quickly identify the drivers behind variances down to specific accounts, departments, or projects.”
“Our turnaround time for monthly budget vs. actuals is so much faster. We are probably saving a couple days’ worth of work because Cube automates so much.”
Cube was founded by Christina Ross, a three-time CFO, and thousands of finance professionals plan with it today. Read verified reviews on G2.
The problem
Where financial planning and modeling breaks down.
Every finance team knows the midnight-before-the-board-meeting version of this work. The difficulty is rarely the finance. It is the plumbing underneath it.
The data arrives late
The first week of every cycle goes to exports and paste-ins. Cube syncs actuals from hundreds of source systems before you open the model.
The model breaks silently
One pasted row shifts a range and the plan is quietly wrong. Cube keeps driver logic governed centrally, outside any single workbook.
Plans live in silos
The revenue plan, hiring plan, and cash forecast disagree by the time anyone compares them. Cube holds all three in one model.
Scenarios take days
By the time the downside case is rebuilt, the question has changed. In Cube, one driver change recalculates every scenario.
Version chaos
Nobody is sure whether v7_final or v8_FINAL went to the board. Cube names, compares, and audits plan versions in one place.
The budget goes stale
The annual plan is obsolete by March and stays on the shelf anyway. Cube reforecasts on live actuals, so the plan moves with the business.
The same plan, a fraction of the work.
The shift
Manual planning vs. the governed model.
The workflow keeps its shape. Every step just stops depending on someone rebuilding it by hand.
| The workflow | Manual planning & modeling | In a governed model |
|---|---|---|
| Data collection | ✗Exports and paste-ins from every system, every cycle | ✓Actuals sync from source systems as they land |
| Model logic | ✗Formulas rebuilt per version; one broken range goes unnoticed | ✓Driver logic defined once, governed centrally |
| Scenario turnaround | ✗Clone the workbook, rework it for days | ✓Change a driver; every case recalculates |
| Version control | ✗v7_FINAL_final2.xlsx, sent by email | ✓Named versions with an audit trail, in one model |
| Collaboration | ✗One locked master file, edited in sequence | ✓Role-based access, parallel contributions |
| Plan vs. actuals | ✗A variance deck assembled by hand each month | ✓Live budget vs. actuals the moment actuals land |
See the difference on your own model.
Bring your current planning workbook. We will show you the same plan running on live data.
The landscape
The six disciplines of financial planning.
Most planning work falls into six connected disciplines. Mature FP&A teams run all of them from one model, so the plans can never disagree with each other.
Driver-based planning
Build the plan on the levers that actually move the business. Read the driver-based planning guide.
Revenue planning & forecasting
Pipeline, bookings, and retention become a revenue plan you can defend. Explore revenue planning and forecasting.
Cash flow forecasting
Know where cash lands before it surprises you. See the cash flow forecasting guide.
CapEx planning
Time large investments against cash and capacity. Read the CapEx planning guide.
Workforce planning
Headcount is most of the cost base. Plan hires, comp, and timing with the drivers they affect. Read the workforce planning guide.
Scenario planning
Base, upside, and downside, side by side. One driver changes and every case recalculates. See the scenario planning guide.
Field notes
Five habits of plans that get used.
Five habits separate teams whose plans get used from teams whose plans get filed. None of them require new software; all of them get easier with it.
Plan on drivers, report on outcomes
Targets are dollars. The plan is the levers that produce them, so a miss points to a cause.
Reforecast on a rolling cadence
A good forecast refreshed quarterly beats a perfect one built annually. Our revenue planning and forecasting guide covers cadence in depth.
Keep revenue, spend, and cash together
One model means the cash flow forecast can never disagree with the revenue plan it depends on.
Version everything, hide nothing
Save every submitted plan as a named version and compare openly. The variance conversation gets shorter.
Close the loop with actuals
Plan vs. actuals is where next quarter's assumptions come from. Review it monthly, in the model that made the plan.
The forecast is never finished. The teams that win reforecast as fast as the business moves.
The evaluation
Six tests for financial planning software.
Six criteria decide whether a planning platform earns its keep. Each is testable in a demo, so test them.
Connects to your real systems
Actuals, pipeline, and headcount should sync from hundreds of source systems. See how Cube's integrations connect your stack.
Meets finance in the spreadsheet
Excel and Google Sheets as first-class surfaces: fetch actuals in, publish plans back, keep your muscle memory.
Driver logic, governed centrally
Logic defined once, versioned, and shared, so no workbook can quietly fork the truth.
Scenarios in minutes
Changing one driver should rebuild every case. If scenarios need a weekend, the tool is a spreadsheet with a login.
AI that shows its work
Agents should propose, explain their inputs, and wait for approval. Meet Cube's FP&Agents.
Security a CFO can sign off on
SOC 2 Type II, SSO, role-based access, and an audit trail are table stakes for the system that holds the plan.
What finance teams ask before they switch.
Plan forward with numbers you can trust.
See a driver-based plan, a live forecast, and a scenario comparison built on your own data.
30 minutes, personalized to your planning cycle and your stack.
