Running out of cash is cited in 38% of startup post-mortems as a primary cause of failure, according to CB Insights, and almost none of those companies died overnight. The cash problem was visible 8 to 12 weeks out. They just were not looking at the right report.

The right report is a 13-week cash flow forecast: a direct-method, week-by-week projection of every dollar entering and leaving your bank accounts over one fiscal quarter. It is the single document that answers the only question that matters when runway gets tight - "what is our bank balance going to be on a specific Friday ten weeks from now?" - and it is increasingly the first artifact investors request when they diligence a bridge round or an extension.

This guide gives you the complete template structure, a step-by-step build process we use across roughly 100 venture-backed clients, the weekly operating cadence that keeps it accurate, and the mistakes that make VCs discount your numbers. It replaces our earlier guide as the definitive version; the original 2025 walkthrough remains available as a related resource.

Why 13 weeks, and why VCs ask for it

Thirteen weeks is one quarter. That window is long enough to see a cash cliff coming with time to act, and short enough that you can forecast from actual invoices, actual payroll runs, and actual contracts rather than assumptions. A monthly budget hides intra-month swings: a company can look fine at month-end and still miss payroll on the 15th because a large customer payment lands on the 28th.

Investors care for a different reason. A 13-week forecast is hard to fake. Your annual model can hide optimism in Q4 hockey sticks, but a weekly cash forecast gets tested against reality every seven days. When a board member or a prospective investor sees a 13-week forecast with a tracked variance history, they learn two things at once: the company's true runway, and whether the finance function can be trusted. During diligence for bridge financings and Series B extensions, this document routinely sits alongside the items in a standard Series B data room checklist.

The macro context makes this sharper in 2026. SaaS Capital's annual survey of more than 1,000 private SaaS companies has documented multi-year declines in median growth rates, which means fundraising timelines stretch and runway math gets less forgiving. Bessemer Venture Partners has been explicit that efficiency metrics now carry as much weight as growth in venture underwriting. A company that manages cash weekly demonstrates exactly the discipline that environment rewards.

The 13-week cash flow template structure

Build this in a spreadsheet before you buy any software. Columns are weeks (W1 through W13, each labeled with its Friday date). Rows follow the direct method: cash receipts, cash disbursements, net flow, and running balance. Here is the full line-item structure:

Three structural rules make this template investor-grade:

Direct method only. Forecast actual cash movements, not revenue minus expenses. Revenue is not cash. A $120k annual contract invoiced upfront with net-60 terms produces zero cash for two months and then one large deposit. Your accrual P&L will never show you that; line 2 of this template exists precisely to capture it.

Weeks are dated, not numbered. Payroll hits on specific dates. Rent hits on the 1st. A "week 7" column means nothing; a "Friday, September 11" column forces you to place every payment on its real calendar date.

Actuals replace forecast every week. The template is a rolling instrument. Each Monday, last week's column converts from forecast to actual, a new week 13 appears at the far right, and the variance between the two becomes a permanent record.

Step-by-step: building your first version

Step 1: Reconcile your starting cash position. Sum every bank account, money market account, and payment-processor balance (Stripe balances in transit are cash you have earned but not received - put them in receipts, not beginning balance). This number must tie to the bank to the dollar. An investor who finds a starting-balance discrepancy stops trusting every number after it.

Step 2: Build receipts from invoices, not revenue. Pull your AR aging. For each open invoice, forecast the collection week using the customer's actual payment behavior, not the due date on the invoice. A customer with net-30 terms who has paid at day 52 for three consecutive quarters is a day-52 payer. For new bookings, only include deals with signatures or verbal commits, and apply contract payment terms to estimate the deposit week. Everything else is upside, and upside does not pay payroll.

Step 3: Schedule disbursements by exact date. Payroll is 60 to 75% of operating cash out for most software startups, and it is perfectly predictable - put each run in its exact week, including the employer tax load. Then work through your vendor list: annual software renewals (these ambush more startups than any other category), rent, insurance, and committed marketing spend. Check last quarter's bank statements for anything recurring you missed.

Step 4: Compute net flow, ending balance, and runway. Lines 16 through 18 are formulas. If ending cash goes negative in any week, you have found the exact date of your crisis, which is the entire point of the exercise. Note that weekly runway math will differ from the burn multiple your board tracks; those measure different things, and the 2026 burn multiple benchmarks for Series A SaaS explain what investors expect on the efficiency side.

Step 5: Build a downside case. Copy the sheet. Delay every uncommitted receipt by four weeks and remove the bottom 20% of expected new-bookings cash. If the downside case breaches your minimum cash threshold (we recommend the greater of two payroll cycles or $250k for most seed and Series A companies), you now have a dated trigger for action: cut spend, draw a credit line, or start the raise. Y Combinator's library has hammered this point since 2022 - default-alive versus default-dead is a weekly, dated calculation, not a vibe.

Step 6: Institute the Monday variance review. Thirty minutes, every week, no exceptions. Enter actuals, compute variance on receipts and disbursements separately, annotate every variance above threshold with a one-line cause, and roll the window forward.

The variance discipline that makes it investor-ready

A forecast without a variance history is a guess with formatting. What separates a CFO-grade 13-week forecast from a founder's spreadsheet is the tracked record of forecast versus actual, week after week. These are the thresholds we hold clients to:

Receipts variance and disbursements variance should be tracked separately, because they fail differently. Disbursements should run under 5% variance almost immediately - you control them. Receipts variance is where the learning happens: every miss teaches you something about a specific customer's payment behavior, and that knowledge compounds into forecast accuracy. We cover how to formalize this into targets in our guide to forecast accuracy KPIs for Series A finance teams.

This weekly rhythm also feeds directly into investor communication. A monthly update that says "cash: $3.1M, forecast accuracy last 13 weeks: 96%" does more for follow-on confidence than a page of narrative. Our monthly investor update template shows where the cash section sits in the broader update.

Five mistakes that make VCs discount your forecast

1. Forecasting revenue instead of collections. The most common error. If your receipts line moves in lockstep with your MRR, an experienced investor knows within a minute that you have not modeled payment terms. David Skok's SaaS metrics work is the canonical reference on why bookings, revenue, and cash are three different curves.

2. Netting receipts against disbursements. A single "net burn" line hides everything diligence wants to see. Gross flows, always.

3. Sandbagging disbursements to flatter runway. Omitting the annual SOC 2 renewal, the D&O insurance premium, and the January payroll-tax reset does not make them not happen. The KeyBanc and Sapphire Ventures SaaS survey is a useful external check on whether your spend assumptions by category are realistic for your stage.

4. Building it once for the board meeting. A 13-week forecast dated six weeks ago, with no actuals entered since, is worse than no forecast. It documents that cash management is theater.

5. No linkage to the operating plan. The 13-week forecast is the near-term expression of your annual plan, not a separate universe. If the quarterly cash trajectory contradicts the model you showed investors, one of them is wrong. Our walkthrough on building an investor-ready Sequoia-style forecast covers how the annual model and the weekly cash view stay reconciled.

When the spreadsheet stops scaling

A spreadsheet is the right starting point, and for many seed-stage companies it is sufficient for a year or more. It breaks when collections span dozens of customers with mixed terms, when usage-based infrastructure spend swings weekly, or when the person maintaining it becomes the single point of failure. At that point teams either adopt dedicated tooling (we compared the main options in our cash flow forecasting software shoot-out) or wire the forecast directly into their systems.

Our own answer is the latter. CFO Advisors is the only fractional CFO firm with an in-house engineering team, and we connect banking, payroll, billing, and AR data into a pipeline that updates the 13-week view continuously and pushes it to founders and boards in Slack. The Monday variance review still happens - the discipline is the product - but nobody spends four hours a week re-keying bank transactions, and there is no six-week month-end lag between reality and the report.

Get this built in two weeks, not two quarters

Most founders who build their first 13-week forecast discover something uncomfortable in week one: a covenant date, a collections gap, an annual renewal nobody budgeted. Finding it now is the win. If you want a hand, this is core work for us - we build the forecast, install the weekly cadence, and fix the underlying billing and collections systems that cause the variance in the first place, rather than reporting the same misses forever. Work with a fractional CFO who treats the 13-week forecast as an operating system, not a board-meeting artifact; the first conversation typically covers your current runway math and where it is most likely wrong.

FAQ

How is a 13-week cash flow forecast different from my financial model?

Your financial model is accrual-based, monthly or quarterly, and assumption-driven; it answers strategic questions over 12 to 36 months. The 13-week forecast is cash-based, weekly, and evidence-driven, built from actual invoices and scheduled payments. It answers one question: what will the bank balance be each Friday for the next quarter. You need both, and they must reconcile.

Who should own the 13-week forecast at a startup?

Someone accountable for the numbers weekly: a fractional CFO, a strong controller, or the founder at pre-seed. Bookkeepers typically cannot own it because it is forward-looking, and the update cadence is weekly rather than monthly. If nobody on the team can commit 30 to 60 minutes every Monday, that is usually a sign it is time to bring in a fractional CFO.

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

Weeks 1 through 4 should land within 5% on total cash flow, since nearly everything is known. Weeks 5 through 13 should be within 10 to 15%, tightening as each week rolls closer. The trend matters more than any single week: accuracy should improve month over month as your collections assumptions absorb real payment behavior.

Do investors actually ask for this document?

Yes, and increasingly early. It is standard in bridge-round and venture-debt diligence, common in Series B processes, and many boards now request it as a standing appendix once runway drops below 18 months. Even when nobody asks, presenting one unprompted signals operational maturity that most competing companies at your stage cannot match.

Should I forecast in a downside case or a base case?

Both, in the same workbook. Operate weekly against the base case, but make decisions - hiring freezes, credit-line draws, raise timing - against the downside case. The downside case is where you set your minimum-cash trigger and attach a date to it.

Can I just use my accounting software's cash flow report instead?

No. Accounting systems report historical cash flow, usually on a lag, and their built-in projections rarely model customer-level payment behavior or exact payroll dates. The 13-week forecast is a forward-looking operating tool. Your accounting system is a source of inputs for it, not a substitute.

Sources

1. CB Insights - Research on top reasons startups fail
2. SaaS Capital - Annual survey and research on private SaaS company growth and retention
3. Bessemer Venture Partners Atlas - Guidance on efficiency metrics in venture underwriting
4. Y Combinator Library - Essays on default alive vs. default dead and startup cash management
5. David Skok, For Entrepreneurs - SaaS Metrics 2.0 on bookings, revenue, and cash timing
6. KeyBanc Capital Markets and Sapphire Ventures - Annual SaaS Survey spend and efficiency benchmarks