13-Week Cash Flow Forecast: Template to Prevent Cash Crises

82% of startups fail because of cash flow problems. Not a bad product, not a weak team — the money runs out before the business hits profitability. Yet most founders just watch their account balance and monthly P&L without ever forecasting cash flows on a weekly basis.

Here’s how to build a 13-week cash flow forecast from scratch: table structure, categories, formulas, scenario analysis, and AI prompts that cut the build time from days to hours.

Why 13 weeks

One quarter. Thirteen weeks covers exactly one financial quarter — short enough for accurate forecasts, long enough to make real decisions.

A monthly forecast hides cash crunches within the month. Revenue might arrive on the 28th while payroll goes out on the 5th. The monthly report looks fine; the bank account’s been negative for two weeks.

Weekly granularity shows the real picture: when money actually comes in and when it goes out. You spot weeks with a negative balance 4–8 weeks before the problem turns critical.

Three key advantages of the 13-week format:

Early detection of cash gaps. Problems show up 4–8 weeks out. That’s enough time to speed up collections from customers, renegotiate expenses, or arrange bridge financing.

Planning discipline. Weekly forecast updates capture variances against plan. After 2–3 months, you’ve built a history that makes future forecasts more accurate.

The language of investors and lenders. A 13-week cash flow forecast is a standard instrument in due diligence. Investors expect it for any funding round. Banks ask for it when evaluating credit lines.

Table structure

The forecast has three blocks: inflows, outflows, and net cash flow.

Columns

14 columns: one for the category and 13 for the weeks. Each week is labeled by its Monday date (or Friday, depending on your accounting cycle). Week 1 is the current week.

| Category           | Wk 1    | Wk 2    | ... | Wk 13   |
|                    | Mar 31  | Apr 7   | ... | Jun 23  |

Block 1: Cash inflows

All sources of incoming cash — not accrual revenue, but actual deposits to your account.

Cash Inflows
├── Customer payments — existing contracts
├── Customer payments — new sales
├── VAT refunds / tax credits
├── Grants / subsidies
├── Investment tranches
├── Loan proceeds
└── Other inflows
TOTAL INFLOWS

Existing contracts are forecast with precision — dates and amounts are known. New sales are a probabilistic estimate weighted by pipeline conversion rates.

Block 2: Cash outflows

All outgoing payments, grouped by degree of obligation.

Cash Outflows
├── Payroll (salaries + taxes)
├── Rent and utilities
├── Subscriptions and SaaS tools
├── Server infrastructure (hosting, cloud)
├── Marketing and advertising
├── Contractors and freelancers
├── Legal and accounting services
├── Tax payments
├── Loan repayments
├── Capital expenditures (equipment)
└── Miscellaneous expenses
TOTAL OUTFLOWS

Payroll and rent are fixed — forecast precisely. Marketing and contractors are discretionary — adjust them when a cash gap threatens. Separating fixed from discretionary expenses matters a lot for scenario analysis.

Block 3: Net cash flow

TOTAL INFLOWS
– TOTAL OUTFLOWS
= NET CASH FLOW (weekly)
+ OPENING BALANCE
= CLOSING BALANCE

The closing balance at the end of week N automatically becomes the opening balance for week N+1. Change one week and it cascades through every week after it.

Formulas and calculations

Weekly net cash flow

Net_CF[n] = Total_Inflows[n] - Total_Outflows[n]

Cumulative balance

Closing_Balance[n] = Opening_Balance[n] + Net_CF[n]
Opening_Balance[n] = Closing_Balance[n-1]
Opening_Balance[1] = Current account balance

Pipeline-to-cash conversion

To forecast inflows from new sales, use a formula that connects the sales pipeline to actual cash:

Expected_Inflow[n] = Σ (Deal_Value[i] × Win_Probability[i])

where the sum is over deals expected to close in week n. Win_Probability comes from historical conversion data at each pipeline stage.

Example: three deals in week 5.

Deal	Amount	Pipeline Stage	Probability	Weighted Amount
A	$10,000	Negotiating	40%	$4,000
B	$5,000	Invoice sent	80%	$4,000
C	$15,000	Qualifying	15%	$2,250
Total				$10,250

Forecasting payments from existing customers

If customers pay on Net-30 but historically average 12 days late:

Expected_Payment_Date = Invoice_Date + Payment_Terms + Avg_Delay

For SaaS with automatic subscription billing, the forecast is nearly deterministic: dates are known, and the only variable is churn.

Subscription_Inflow[n] = MRR × (1 - Weekly_Churn_Rate)^n

Minimum cash floor

Conservative approach: a minimum of 4 weeks of fixed expenses.

Cash_Floor = 4 × Weekly_Fixed_Costs

For a startup with $20,000 in weekly fixed costs, Cash Floor = $80,000. Any week where the forecast balance drops below this gets flagged in red.

Step-by-step build

Step 1: Gather the source data

• Current balance across all bank accounts
• List of recurring payments with dates (payroll, rent, subscriptions)
• Open accounts receivable (AR) with dates and amounts
• Open accounts payable (AP) with dates and amounts
• Sales pipeline with stages and deal values
• Historical pipeline conversion rates by stage

Step 2: Fill in the fixed lines

Start with what you know precisely: payroll, rent, subscriptions, loan payments. Fill these in for all 13 weeks.

Step 3: Distribute AR and AP across weeks

Spread outstanding receivables across the weeks when payment is expected (accounting for historical delays). Spread outstanding payables across the weeks when payment is planned.

Step 4: Add probabilistic inflows

New sales from the pipeline, weighted by probability. Update weekly as deals move through the funnel.

Step 5: Calculate totals and check the cash floor

Apply the formulas. Flag any weeks where the balance drops below the Cash Floor. If there aren’t any, the base case is healthy.

Step 6: Weekly updates

Every Monday (or Friday):

1. Replace last week’s forecast with actuals
2. Add week 14 to the end (rolling window)
3. Update pipeline and deal probabilities
4. Log the actual vs. forecast variance

Turn this into a repeatable standard operating procedure so the update doesn’t depend on one person’s memory.

Accumulated actual vs. forecast variances are the most valuable artifact. After 8–10 weeks, they reveal systematic errors: are inflows consistently overestimated, are certain expense categories underpredicted, which lines are the least predictable.

Scenario analysis: base / optimistic / pessimistic

One forecast is a point. Three scenarios are a range.

Base case

Current trends continue. Pipeline conversion matches the last 3 months. Expenses grow as planned. Churn holds at current levels.

Optimistic case

Pipeline conversion x 1.2. Average deal size x 1.1. Churn x 0.8. Expenses unchanged. Use this for planning growth investments.

Pessimistic case

Pipeline conversion x 0.6. Two largest customers delay payment by 4 weeks. One churns. Expenses unchanged — you can’t cut fixed costs immediately.

The pessimistic case answers: “How many weeks can the company survive if everything goes wrong?” If the answer is fewer than 8 weeks, you need to act now.

Scenario coefficient table

Parameter	Pessimistic	Base	Optimistic
Pipeline conversion	x0.6	x1.0	x1.2
Avg. new deal size	x0.9	x1.0	x1.1
Payment delay (days)	+14	0	-5
Churn rate	x1.3	x1.0	x0.8
Marketing spend	x1.0	x1.0	x1.15

Common mistakes

Confusing revenue with cash. An issued invoice isn’t money in the bank. A SaaS company with Net-30 terms can have $100K MRR and an empty account if customers pay late while expenses go out on time.

Ignoring seasonality. B2B sales dip in December–January and August. If you build a forecast in January and extrapolate January data across the entire quarter, you’ll underestimate inflows.

Counting pipeline at face value. $500K in pipeline doesn’t equal $500K in cash. Probability-weighting is mandatory — without it, the forecast consistently overstates inflows. If you haven’t mapped your ideal customer profile, pipeline probabilities are guesswork.

Forgetting one-time expenses. Annual licenses, tax payments, bonuses — costs invisible in monthly P&L that burn through cash in specific weeks.

Not updating the forecast. A 13-week forecast loses its value if you only update it monthly. Weekly updates are the minimum.

AI prompts for building the forecast

Prompt 1: Generate the table structure

Create a 13-week cash flow forecast structure for [business type, e.g.: B2B SaaS with $500K ARR, 12-person team, office on a lease].

Requirements:
- Grouping: Inflows, Outflows, Net Cash Flow
- Inflows split into: existing contracts, new sales, other
- Outflows split into: fixed, discretionary, one-time
- Add rows: Opening Balance, Closing Balance, Cash Floor
- Format: table with 14 columns (category + 13 weeks)
- Include formulas for calculated rows

Return the table as CSV, ready to import into Google Sheets.

Prompt 2: Cash flow risk analysis

Analyze the following 13-week cash flow forecast and identify risks:

[paste table or description of key metrics]

Check for:
1. Weeks where Closing Balance falls below Cash Floor ($[amount])
2. Inflow concentration (dependence on 1–2 large customers)
3. Timing gap between major outflows and inflows
4. Net Cash Flow trend — improving or declining
5. Sensitivity to a 2-week and 4-week payment delay

Response format:
- Top 3 risks with specific weeks identified
- For each risk: trigger, consequence, action

Prompt 3: Scenario analysis

Based on the following data, build three 13-week cash flow scenarios:

Base case data:
- MRR: $[amount]
- Pipeline: $[amount] (average conversion [X]%)
- Fixed expenses: $[amount]/week
- Discretionary expenses: $[amount]/week
- Current balance: $[amount]
- Average payment delay: [N] days

Coefficients:
- Pessimistic: conversion ×0.6, payment delay +14 days, loss of 1 major customer
- Base: current trends
- Optimistic: conversion ×1.2, delay -5 days, churn -20%

For each scenario:
- Closing Balance by week
- Week where balance falls below Cash Floor (if applicable)
- Runway in weeks

Prompt 4: Categorize expenses from bank statement

Categorize the following transactions into cash flow forecast categories (Payroll, Rent, SaaS Tools, Infrastructure, Marketing, Contractors, Legal, Taxes, CapEx, Other):

[paste transaction list]

For each transaction, identify:
- Category
- Type: fixed / discretionary / one-time
- Frequency: weekly / monthly / quarterly / one-time
- Which forecast week the next payment falls in

Return a table sorted by category.

These prompts work with any LLM, but models with long-context support — Claude, GPT-4o, Gemini — give the best results since a 13-week table can contain hundreds of cells. If you’re running multiple prompts in production, a prompt management system helps track versions and results.

Connection to unit economics

A cash flow forecast doesn’t replace unit economics — it fills a different gap. Unit economics answers “is the business profitable at the customer level.” Cash flow forecast answers “will there be enough money in the account to survive until profitability.”

Example: LTV/CAC = 3.0 — a strong signal — but Payback Period = 14 months. If the company is aggressively acquiring customers, the acquisition costs land today while the return shows up in a year. The cash flow forecast tells you whether there’s enough runway to cover those 14 months.

Weekly_CAC_Spend = New_Customers_Target × CAC

This line goes into Outflows, and inflows from new customers go into Inflows with a lag equal to the Payback Period. The gap between acquisition spend and first cash return is the main source of cash gaps for high-growth SaaS companies.

Tools

Google Sheets

The minimum viable tool. Cascading calculation formulas, conditional formatting for Cash Floor violations, separate tabs for scenarios. Share it with the team and investors without any extra tooling.

Key functions:

• SUMIF for category aggregation
• Conditional formatting: red background for weeks below Cash Floor
• Data Validation for expense type (fixed/discretionary/one-time)
• Separate tabs: Base, Pessimistic, Optimistic, Actuals

Notion / Airtable

Useful for tracking AR/AP with links to customers and projects. Doesn’t replace the spreadsheet, but adds structured deal and payment data on top.

Dedicated tools

For companies with revenue above $1M/year: Float, Pulse, Agicap. They plug into banks and accounting systems, pulling in transactions automatically. Cost: $50–500/month.

FAQ

How often should I update a 13-week cash flow forecast?

Weekly. Every Monday or Friday, replace last week’s forecast with actuals, add a new week 14, and update pipeline probabilities. Monthly updates defeat the purpose — the whole point is catching problems weeks before they hit.

What’s the difference between a cash flow forecast and a P&L statement?

A P&L shows accrual-based revenue and expenses. A cash flow forecast shows when money actually moves in and out of your bank account. You can be profitable on the P&L and still run out of cash if customers pay late.

How accurate should a 13-week cash flow forecast be?

Weeks 1–4 should land within 10–15% of actuals. Weeks 5–8 within 20–25%. Weeks 9–13 are directional — the goal is spotting trends, not precision. Track actual vs. forecast variances weekly to calibrate over time.

Can I use a 13-week cash flow forecast for fundraising?

Yes. Investors and lenders expect it during due diligence. It demonstrates financial discipline and shows you understand your cash dynamics. Pair it with a pessimistic scenario to show you’ve stress-tested the numbers.

Where to start

The first version of the forecast doesn’t need to be perfect. It just needs to exist.

Today. Open Google Sheets. Create 14 columns. Enter the current balance. Fill in fixed expenses for 13 weeks. That’s 30 minutes, and it’ll already show your basic burn rate.

This week. Add AR and AP. Spread outstanding invoices across weeks. Add pipeline with probabilities. Calculate the Cash Floor. Highlight the problem weeks.

In a month. You’ll have 4 actual vs. forecast data points. Check accuracy by category. Add the pessimistic scenario. Start weekly Friday updates.

A 13-week cash flow forecast isn’t an investor report. It’s an operational tool that turns “I think we have enough money” into specific numbers, week by week. Companies that run one see problems a month before they hit. The rest find out when the account hits zero.