There is a moment in every Series A diligence process when an investor asks to see the balance sheet. Most founders open the model, scroll past the P&L, and find something they haven't looked at in months — a tab with numbers that don't match anything, a cash balance that disagrees with the bank, and equity that mysteriously balances because someone put a plug in column F.
That moment is recoverable. But it costs trust, and trust is the only currency that matters in a fundraise.
What "closing" actually means
A three-statement model closes when the following conditions hold simultaneously:
- The cash on the balance sheet equals the ending cash on the cash flow statement. Every dollar in, every dollar out, every period.
- Retained earnings on the balance sheet equals cumulative net income from the P&L. No gaps, no plugs.
- The balance sheet balances. Assets = Liabilities + Equity. Not approximately. Exactly.
If any of these fail, the model is not a three-statement model. It is a P&L with decoration.
Why most startup models don't close
The failure usually starts with the cash flow statement. Most founders build a direct cash flow — they list cash inflows and outflows — rather than an indirect one derived from the P&L and balance sheet movements. The direct method looks right but creates a parallel universe that drifts from the balance sheet over time.
The indirect method is harder to build and harder to explain, but it is the one that closes. It starts with net income and works backward:
- Add back non-cash charges (depreciation, amortization, stock compensation)
- Adjust for changes in working capital (accounts receivable, deferred revenue, accounts payable)
- Add investing activities (equipment purchases, software capitalization)
- Add financing activities (debt, equity raises)
- The result is the change in cash — which must equal the ending cash on the balance sheet
The working capital trap
This is where startup models most often break. SaaS companies collect cash before they recognize revenue. That gap — deferred revenue — is a liability on the balance sheet. When deferred revenue increases (you collected more than you recognized), that is a source of cash. When it decreases, it is a use.
If your model doesn't have a deferred revenue schedule that feeds both the P&L (revenue recognition) and the balance sheet (current liability), your cash flow will be wrong. Always.
The same logic applies to accounts receivable. If you invoice in December but collect in January, your P&L shows December revenue, your cash flow shows January cash. The AR balance is the bridge. A model that doesn't track AR will overstate cash in any month where billings precede collections.
What investors actually check
When a sophisticated investor opens your model, they run three checks before reading any of the assumptions:
- Does cash on the balance sheet match cash on the cash flow statement?
- Does the balance sheet balance?
- Does the equity section reflect cumulative net income, less any distributions?
If all three pass, they proceed to the assumptions. If any fail, they open a new tab and begin drafting the "we'd like to continue our diligence" email — which is the polite version of no.
Building it the right way
Start with the P&L. Get revenue, COGS, and operating expenses right. Then build a balance sheet with real schedules: AR aging, deferred revenue waterfall, fixed asset rollforward, debt schedule. Every line on the balance sheet should be supported by a schedule — not a formula that says "last period + new activity."
Once the balance sheet has real schedules, the cash flow statement writes itself. It is the mathematical consequence of how the balance sheet changed. If it doesn't match the bank, you have a schedule wrong — not a formula wrong. Find the schedule.
The forty hours founders spend rebuilding models before a fundraise is almost always spent on this: retroactively adding the infrastructure that should have been there from the start. Build it at the start.