The IFM (Integrated Financial Model)

The IFM is the working instrument of the Fourth Statement — a single integrated model that combines historical reporting and forecasting for a small business.

One spreadsheet, one system: the income statement, balance sheet, operating drivers, and forecast all connected, all reconciled back to the books, updated weekly.

The Architecture

An IFM is built from a standard template (~25 tabs) that grows as the business is customized:

• Core model — monthly grain (the IFM tab) paired with a weekly grain twin (the MMM)
• The SEQ — business-specific drivers at the top: customers, orders, conversion, AOV, acquisition cost
• Quality layer — bookkeeping quality checks, an “Ask” queue for unrecognized transactions, AR/AP aging
• Source tabs — raw data pulled from the accounting system, so every number in the model traces to the books
• Error checker — validates the accounting equation and cross-checks every section. Every value should be zero. If it isn’t, you’ve found a problem worth knowing about.

The Actuals Boundary

The defining feature: a moving line between history and forecast. Everything left of the boundary is real, reconciled data. Everything right of it is a forecast built bottom-up from drivers — ad spend → customers → orders → revenue — not a growth rate pasted onto last year.

When the boundary goes stale, the model silently degrades into a historical reporting tool. In our experience reviewing dozens of models, this is the single most common failure — a model that looks like a forecast but no longer forecasts anything.

Forecast Methods

Each line item gets an explicit method — no hidden assumptions:

1. Growth Rate — compounding from a base
2. % of Revenue — scales with the top line
3. Flat Line — constant
4. Manual Entry — deliberate human judgment
5. Scheduled / Lookup — known future events (rent steps, loan payments)
6. Derived — computed from other drivers

The Review Sequence

Reviewing an IFM follows a fixed order, because each step depends on the one before it:

1. Bookkeeping quality — can we trust the data at all?
2. Actuals boundary — is the model current?
3. Error checker — does the math hold?
4. Revenue honesty — does the forecast respect seasonality and reality?
5. Cost survival — do expenses scale sensibly?
6. Customer math — do the unit economics support the revenue story?
7. Balance sheet reality — does cash actually work?

Only after steps 1–3 pass does the forecast even deserve attention. A beautiful forecast on top of broken books is noise dressed as signal.