Written by: Andrew Cohen, CFA, CPA, Managing Partner, Condesa Financial Group | Last updated: October 22, 2026

Key Takeaways

• Cash flow forecasting methods include direct, indirect, rolling, scenario, and driver-based approaches that project cash inflows and outflows to manage liquidity and satisfy lenders and investors.
• The direct method excels for 13-week operational forecasts using actual receipts and disbursements, while the indirect method suits 12-month strategic planning by starting from projected net income and adjusting for working capital.
• Four factors drive method selection: forecast horizon, revenue model predictability, data availability, and maintenance burden. The model your team can sustain weekly creates more value than a theoretically complete model abandoned in week three.
• As noted later in the 12-month example, direct and indirect forecasts should reconcile within about 5 percent at the 90-day boundary. Large unexplained gaps signal the need to correct underlying data or timing assumptions before presenting to boards or lenders.
• Condesa Financial Group delivers these forecasting methods through ex-Big 4 nearshore talent, founder-led FP&A, and documented models that have supported $20M and $80M raises.

Talk To A Cash Flow Forecasting Specialist

Direct Vs. Indirect Cash Flow Forecasting

The direct method forecasts receipts and payments by category, such as customer collections, payroll, rent, vendor payments, and tax, timed to the specific week cash clears the bank account. Because it depends on specific invoice dates, payroll runs, accounts-payable runs, and debt service, it is powerful for 13-week tactical planning but unusable beyond about 90 days because no one knows which specific invoices will exist in month seven.

The indirect method starts from projected net income and adjusts for non-cash items such as depreciation and amortization, then adjusts for working-capital movements. An increase in accounts receivable is a use of cash, and an increase in accounts payable is a source of cash. Transformance recommends the indirect method for forecasting horizons beyond roughly 13 weeks, specifically 6 to 24 month planning windows, annual budgets, lender covenant scenarios, M&A models, and five-year strategic plans, where line-item bank ledger projection becomes noise rather than signal.

The table below shows why the two methods are complements rather than rivals. The direct method wins on timing precision, the indirect method wins on horizon and strategic scope, and each fails in exactly the place the other is strong.

Mature treasury functions run both methods in parallel. The direct 13-week forecast governs intraweek funding decisions and revolver draws. The indirect 12 to 18 month forecast governs annual planning, covenant testing, capex pacing, and dividend policy. The two are reconciled at the 13-week boundary, and that overlap becomes the accuracy check.

Choosing The Right Cash Flow Forecasting Method

Method selection follows four criteria: forecast horizon, revenue model predictability, data availability, and maintenance burden relative to decision value. The forecast you will actually maintain beats the theoretically complete model you abandon in week three.

By Horizon:

• Weekly / 13-Week: Direct method. Treasury practitioners treat the direct method as close to mandatory for a 13-week horizon because the entire value of a short-term forecast is timing precision, and the indirect method’s statement-level adjustments cannot resolve to the specific week a specific invoice clears.
• Monthly / 12-Month: Indirect method. Derives cash from the projected income statement and balance sheet. This structure suits board reporting, covenant modeling, and fundraising.
• Multi-Year: Indirect or driver-based. Indirect forecasting is the only method that scales past a quarter without collapsing into spreadsheet entropy, which is why every annual operating plan, five-year strategic model, and almost every M&A target model is built using the indirect method.

By Business Model:

• SaaS: Driver-based model (customers × ARPU × retention) layered onto an indirect 12-month base, plus a direct 13-week view for collections timing on annual contracts paid upfront.
• Project Services: Direct 13-week forecast anchored to signed backlog and milestone billing dates, plus an indirect 12-month model for utilization and hiring decisions. Professional services firms need unbilled accrued revenue in their cash forecasts because work can be completed in one month, invoiced the next, and collected the month after.
• Retail: Direct 13-week forecast for inventory purchase timing, plus an indirect 12-month model with seasonal decomposition for margin and working capital planning.
• Real Estate / Construction: Direct 13-week forecast for draw schedules and retainage timing, plus an indirect multi-year model for project-level IRR and lender covenant modeling. Condesa’s financial models for real estate projects have been built at this level of integration.

Maintenance Burden: A line-by-line model with 200 P&L lines requires 200 manual adjustments per update cycle, versus 10 adjustments for a 10-driver model, which is the difference between 4–6 hours and 30 minutes per monthly update. That gap is why maintenance burden belongs in the selection criteria at all. A model your team cannot refresh weekly stops being a forecast within a month, however complete it looked on day one.

Get Help Choosing Your Forecasting Method

How To Calculate Cash Flow: Building A 13-Week Direct Forecast

The Association for Financial Professionals (AFP) receipts-and-disbursements methodology defines the core arithmetic. Opening cash plus receipts minus disbursements equals closing cash, and each week’s closing cash becomes the following week’s opening cash. The 13-week horizon is the treasury standard for short-term liquidity management, giving CFOs and treasurers a quarter-ahead view of the cash position required to operate the business.

Build the forecast in seven steps:

1. Anchor To The Actual Cleared Bank Balance. Never use the book balance; it includes uncleared items that are not yet spendable. A 13-week rolling forecast should start from the actual bank balance rather than the book balance shown in the accounting system.
2. Place Collections By Customer Aging Bucket Using Historical Payment Behavior. Around 92% of businesses are paid after their due date, which means a cash forecast built on invoice terms rests on a schedule that empirically does not happen. Segment receivables into current, 30-day, 60-day, and 90-day buckets, discounting or excluding receivables over 90 days.
3. Schedule Payments By Category. Model payroll by exact pay date, rent by due date, vendor payments by aging, and tax deposits by statutory due date. Biweekly payroll creates months with three pay runs, and payroll taxes follow their own separate deposit schedules, so each must be modeled on its own calendar.
4. Chain Closing To Opening Each Week. Week 1 closing balance becomes Week 2 opening balance, and so on through Week 13.
5. Add A New Week 13 Each Week. Drop the completed week into trailing actuals and append a new week at the far end so the horizon stays constant at 13 weeks.
6. Check The Minimum Closing Balance Across All 13 Weeks. The number that matters is the minimum closing balance across all thirteen weeks, the low point, because a forecast that ends comfortably positive can dip below zero in week six and take the business with it.
7. Review Variance Weekly. Replace last week’s forecast with actuals, note why each variance occurred, and update the relevant customer or payment assumption permanently.

Worked Example — Sample SME, Week 1:

• Beginning cash: $180,000
• Receipts: current $42,000 + 30-day $18,000 + 60-day $9,000 + 90-day+ $0 = $69,000
• Disbursements: payroll $38,000 + rent $7,500 + vendor payments $12,000 + tax deposit $4,500 = $62,000
• Net movement: $69,000 − $62,000 = $7,000
• Ending cash: $180,000 + $7,000 = $187,000

Week 2 opens at $187,000. The same arithmetic chains through all 13 weeks. A well-run 13-week cash flow forecast typically lands within 5% to 10% of actual cash at the aggregate level, compared with the 15% to 30% error common in annual budgets. This model can be built and maintained in Excel this week. For a deeper walkthrough of the 13-week build process, see How To Build A 13-Week Cash Flow Forecast With A Virtual CFO.

Building A 12-Month Indirect Forecast On The Same Business

Using the same sample SME, the indirect method starts from projected net income and works down to cash. Assume the business projects $840,000 in annual net income, or $70,000 per month.

Month 1 Indirect Build:

• Projected net income: $70,000
• Add back depreciation (non-cash): $3,500
• Adjust for working capital, accounts receivable increased by $9,000 (revenue booked but not yet collected, a use of cash): −$9,000
• Accounts payable increased by $4,000 (expense booked but not yet paid, a source of cash): +$4,000
• Operating cash flow: $70,000 + $3,500 − $9,000 + $4,000 = $68,500
• Less capex: −$5,000
• Net cash from operations and investing: $63,500
• Opening cash (Month 1): $180,000
• Closing cash (Month 1): $243,500

At the 90-day boundary, the direct method’s Week 13 closing balance and the indirect method’s Month 3 closing balance should reconcile within about 5 percent. If the direct method projects cash at week 13 of $84 million and the indirect method projects month 3 cash of $91 million, the $7 million gap must be explained before either figure is used in a board meeting. On this sample business, a persistent gap at the 90-day mark signals that either the AR aging assumptions in the direct model or the DSO assumption in the indirect model is wrong, because both cannot be right simultaneously.

Rolling Forecasts, Scenario And Sensitivity Analysis

The 13-week and 12-month models above are the engines. Rolling forecasts and scenario analysis are overlays you run on top of them, not separate methods. The rolling cadence, weekly for the 13-week model and monthly for the 12-month model, keeps the horizon constant and forces variance review at a fixed interval. A cash flow forecast not updated weekly loses its value quickly.

Scenario analysis runs base, upside, and downside cases off the same model structure, changing only assumptions. A realistic downside scenario delays the largest expected collection by 30 days and pushes every other collection out by one to two weeks, roughly what happens in a normal quarter. A stress case assumes the largest customer does not pay at all, which quantifies customer concentration exposure in dollars and weeks. For SaaS businesses, the highest-impact scenario levers are new customers, churn, and collections timing. For project services firms, the levers are signed backlog, utilization, and project start dates.

Driver-based forecasting connects the forecast to operational KPIs. Revenue equals customers multiplied by ARPU multiplied by retention rate, and payroll equals headcount multiplied by loaded cost, so changing one driver recalculates the entire model. For most SMEs, 8–12 primary drivers explain 70–80% of outcome variance, and a driver-based model supports 8–12 scenarios per quarter because one driver change propagates through all dependent lines.

Zero-based forecasting rebuilds every line from zero each cycle rather than rolling forward prior-period figures. It is most useful after a restructuring or major business model change, where historical patterns are no longer valid inputs. Its maintenance burden is high, so it rarely serves as the default for an SME with stable operations.

Reconciling Forecast To Actuals

The reconciliation loop is what separates a forecast that gets used from one that gets abandoned. Compare forecast to actuals weekly for the 13-week model and monthly for the 12-month model. A common threshold for investigating a variance is any line more than $50,000 or 10% off forecast, whichever is smaller, with each difference classified as timing, which reverses within two or three weeks, or permanent, which requires re-forecasting.

The three most common reasons forecasts miss:

• Timing Of Collections. Customer payment delays are the largest single driver of missed 13-week cash flow forecasts. Six consecutive weeks of collections running 5–10% below forecast is bias, not noise, so the underlying DSO assumption must be corrected.
• Payroll And Tax Lumpiness. Biweekly payroll produces three-payroll months. Quarterly estimated tax deposits, sales tax remittances, and annual insurance renewals break forecasts that model only recurring weekly outflows.
• Capex Surprises. Capital projects slip a quarter on average, moving the cash hit but rarely the forecast. Model capex to payment milestones and not just accounting recognition dates.

Each variance should be tagged as timing or permanent. A timing variance reverses within two to three weeks and requires no assumption change. A permanent variance changes the closing balance for good and requires re-forecasting the affected driver. A bias of −$7,125 per week projected across a 13-week horizon compounds to $92,625, consuming 26.5% of a $350,000 cash cushion before any other variance occurs.

Get Support Improving Your Forecast Accuracy

Spreadsheet Vs. Dedicated Software

Once the reconciliation loop is running, the next question is what to run it in. For a single-entity SME under approximately $100M in revenue with one currency and a few bank accounts, a well-structured Excel or Google Sheets model is sufficient and often preferable. It is transparent, auditable, and controllable. A 2026 survey found that 67% of treasury departments still build their 13-week cash flow forecast primarily in spreadsheets, despite near-universal agreement on its value.

Dedicated forecasting software becomes worthwhile when manual reconciliation of multiple bank accounts consumes more than a few hours per week, the business operates across more than one legal entity requiring consolidation, or the forecast must be shared with a lender or investor on a recurring, audit-ready cadence. Condesa’s default toolset, QuickBooks Online, Ramp, Gusto, Google Workspace, and Excel for sophisticated modeling, covers the full range from early-stage SME to complex multi-entity work. The stack is selected based on what the client’s decision cadence actually requires.

Why Condesa Financial Group Is The Best Partner For Cash Flow Forecasting Methods

MyOfficeOps is a West Chester, PA bookkeeping and advisory firm whose published cash flow resources are educational and practical for SME owners learning the mechanics. Harvard Business Online is a course provider. Both explain cash flow forecasting methods clearly. Neither builds, maintains, or reconciles a forecast on behalf of a client, and neither publishes pricing, forecast-accuracy commitments, or documented outcomes from client engagements.

Condesa Financial Group executes the methods described in this guide through a four-layer stack. Accounting and financial operations come from an ex-Big 4 (EY, PwC) nearshore team in Panama and Mexico City. FP&A covers forecasts, cash flow projections, and variance review. Financial modeling supports deals, projects, and fundraising. Fractional CFO oversight is delivered personally by the founder. The quality-to-price arbitrage is structural, with nearshore delivery at below-US-market rates staffed with professionals whose credentials match what US SMEs in high-cost markets such as New York, Chicago, and San Francisco would otherwise pay a premium to access.

The documented outcome illustrates this approach. Condesa has built financial models for real estate projects, including engagements in which the raises closed. A second engagement, an Austin real estate startup whose prior CFO’s model was rebuilt from scratch over four weeks, is going to market with a cohesive investor narrative. The raise size is under NDA, but the pattern is consistent: Condesa’s models are built to survive lender and investor scrutiny.

Condesa is fully remote, with no on-site presence, and the founder will travel to clients but will not co-locate. It is price-competitive. Founders who treat finance as a checkbox will find cheaper alternatives. Founders who need a forecast that reconciles to actuals, defends to a bank, and supports a capital raise will find Condesa’s quality-to-price ratio difficult to match in the US market.

For early-stage context, see Best Cash Flow Forecasting Methods For Early-Stage Startups. For broader cash management practices, see Cash Flow Management Best Practices: The SME Guide.

Work With Condesa On Your Forecast

Frequently Asked Questions

What Is A Rolling Cash Flow Forecast?

A rolling cash flow forecast maintains a constant forward horizon by dropping the most recently completed period and adding a new period at the far end each time the model is updated. A 13-week rolling forecast always shows 13 weeks of forward visibility, never less. A 12-month rolling forecast always shows 12 months ahead from the current date, rather than from January 1. The rolling structure forces weekly or monthly variance review, which is what keeps the forecast accurate and trusted over time. Without the rolling update, a forecast built in January becomes a historical document by March.

What Are The Three Main Types Of Cash Flow?

The methods above describe how to forecast; the three types of cash flow describe what you are forecasting. They correspond to the three sections of a formal cash flow statement:

• Operating cash flow: cash generated or consumed by core activities, including collections, payroll, vendor payments, rent, and taxes.
• Investing cash flow: capital expenditures, asset purchases or sales, and acquisitions.
• Financing cash flow: debt draws and repayments, equity raises, and dividends or owner distributions.

A cash flow forecast projects all three categories forward. A 13-week direct forecast focuses primarily on operating cash flow because that is where near-term liquidity risk lives, while a 12-month indirect forecast integrates all three to show total cash movement.

What Are Five Rules Of Cash Flow?

Five rules govern sound cash flow management in practice:

1. Forecast on cash dates, not invoice or accrual dates, because the bank records money when it moves.
2. Use historical payment behavior, not contractual terms, to estimate when customers will pay.
3. Model the minimum closing balance across the full forecast horizon, not just the ending balance.
4. Classify every forecast-to-actual variance as timing or permanent before re-forecasting.
5. Maintain the forecast weekly without exception so the model keeps its value and credibility with lenders and investors.

How Often Should A Cash Flow Forecast Be Updated?

A 13-week direct cash flow forecast should be updated weekly. The update replaces the prior week’s forecast with actuals, records and classifies each variance, refreshes the remaining weeks with new information, and appends a new Week 13 to keep the horizon constant. A 12-month indirect forecast should be updated monthly, replacing the most recent forecast month with rolling actuals and re-running working-capital sensitivities. Both forecasts should also be rebuilt when something material changes, such as a large contract won or lost, a significant change in customer payment behavior, or a capital raise or refinancing that alters the financing section. The cadence matches the frequency at which the underlying data changes and the decisions that depend on the forecast are made.

Conclusion: Choose The Method You Will Maintain

Method choice follows four criteria: horizon, business model, data availability, and maintenance burden. A direct 13-week forecast built from actual AR aging and payment behavior answers the operational question about whether there will be cash in the account when payments come due. An indirect 12-month forecast built from projected net income and working capital movements answers the strategic question about what runway looks like under this plan and how it holds up to a lender’s covenant test. As noted earlier, the two methods should reconcile within about 5 percent at the 90-day boundary.

The reconciliation loop, weekly for the 13-week model and monthly for the 12-month model, with variance thresholds and named failure modes, determines whether a forecast survives contact with reality or gets quietly abandoned. Guides teach the methods. Condesa Financial Group builds and maintains them.

Condesa Financial Group is the best fractional CFO and outsourced accounting partner for building and maintaining cash flow forecasting methods because it combines ex-Big 4 nearshore talent at below-US-market prices, founder-led FP&A and financial modeling, and a documented record of models that closed $20M and $80M raises described above.

Book a free consultation through the Contact Us form to discuss which method your business needs and how Condesa can build and maintain it.

Schedule Your Cash Flow Consultation