Non-Recurring Cash Flows: How I Handle Them in a 13-Week Forecast

One-off cash events — a legal settlement, an asset sale, a tax refund — can move large sums in a single week. Without a structured approach, they get missed, misclassified, or buried in operating cash flows — and when timing shifts or negotiated amounts change, the surprise lands too late to act on.

Non-recurring cash flows are often the most uncertain items in a short-term forecast. They are not part of the day-to-day business cycle, they may involve multiple stakeholders, and their timing frequently shifts. Handled correctly, they add transparency. Left unmanaged, they create sudden surprises — in either direction — that undermine trust in the forecast.

What non-recurring cash flows are

A non-recurring cash flow is cash that enters or leaves the business only once, or very rarely. It does not form part of the normal operating cycle. Examples include proceeds from selling equipment, a one-time legal settlement payment, or a special tax refund.

Three distinctions matter in practice:

Non-recurring vs. irregular recurring. Some cash flows happen again and again but at uneven intervals or variable amounts — for example, annual maintenance spending that varies in cost and timing. These are still recurring, because they are expected to continue. A non-recurring item is truly one-off and unlikely to repeat.

Non-recurring vs. accounting adjustments. Write-offs, depreciation, and non-cash charges move numbers on paper but involve no actual cash movement. The forecast covers only real cash that enters or leaves the bank account.

Non-recurring vs. operating cash flows. Normal sales, monthly rent, payroll, and routine supplier payments belong in operating cash flows. They repeat predictably and form the core of the business. Non-recurring items sit separately so the operating picture remains clean.

In every forecast, non-recurring cash flows are listed in their own section with clear descriptions, amounts, and timing. This keeps the operating section uncluttered and gives all readers — management, lenders, and owners — an accurate view of both dimensions.

Finding them

The first practical task is building a complete list of every upcoming non-recurring event. Missing even one significant item is the main reason forecasts surprise people.

The approach combines conversations with document review. The departments most likely to have relevant information are finance, legal, HR, procurement, operations, and sales. Each is asked directly about any unusual cash events expected in the coming 13 weeks — or further out if payment timing is uncertain.

Non-recurring cash flows by department

Finance and treasury — asset sales or purchases in progress, special financing, unusual loans or investments, one-time tax payments or refunds from audits or special filings.

Legal — pending litigation settlements based on expected timelines and outcomes, ongoing disputes that may require payment, guarantees that might be called in.

HR — planned programs for staff reductions with estimated costs and payment dates, severance or retention bonuses with specific payment terms, one-off executive compensation arrangements.

Procurement — exceptional supplier payments outside normal terms, non-recurring deposits or advance payments, contract dispute resolutions.

Sales and operations — large one-off customer receipts, customer refunds from disputes or returns, contract termination payments in either direction, insurance claims expected or in progress, major facility changes.

The identification process runs at the start of each forecast engagement and repeats every week during the monitoring phase. New events surface regularly, and the earlier they are captured, the more accurately they can be timed and estimated.

Classification

When there are only a few non-recurring items, classification adds little value. Each item simply gets its own clearly labelled row with a short note explaining what it is. That is sufficient.

When the volume grows, grouping becomes useful — not for its own sake, but to keep the forecast readable and easy to discuss. The grouping logic follows whatever makes most sense for the specific company and audience:

• By nature of the event — the same categories used in the income statement or management accounts, so readers work with familiar labels.
• By department — useful when several departments each own multiple items and accountability needs to be clear.
• By client requirement — some lenders or investors require operating / investing / financing classification. When that is the case, the forecast follows their format.

Events of the same type and low individual value are grouped into a single line to keep the model readable. A new separate category is created only when an event is large, unique, or needs specific attention from management or lenders.

The default is to keep it simple. Structure is added only when it genuinely helps the reader navigate the material.

Forecasting each event

Each identified and classified event is then forecast on four dimensions: timing, amount, probability, and ownership.

Timing

The question is not when an event is expected in general terms, but when cash will actually move through the bank account. These are often different. A signed settlement agreement may take three to four weeks to result in an actual payment. An asset sale may close in week five but with proceeds arriving in week six or seven.

The milestone that triggers payment is identified explicitly — whether that is deal closing, invoice approval, court ruling, or board sign-off. Timing is always assessed for slip (delay) risk (probability). Legal settlements, supplier negotiations, and contract terminations frequently move between weeks, and the forecast must reflect this rather than assume the original date holds.

Amount

The amount is estimated using available documentation: contracts, quotes, legal advice, or direct input from the responsible team. Some parts are fixed — for example, an agreed settlement figure or a penalty clause with a stated amount. Other parts are variable, such as final legal fees or costs that depend on how negotiations conclude. Where variability exists, a conservative estimate is used and the range is noted in the assumptions.

Probability

Not every identified event will happen within the forecast window. Each item is assigned a confidence level — high for board-approved items with signed agreements, medium for events in active negotiation, low for potential claims or contingent items. Low-probability events are included in the forecast only when their potential cash impact is material. When they are included, they are clearly marked and typically shown in a separate scenario rather than the base case.

Ownership

Each forecast line for a non-recurring item has a named owner — usually the department head or project lead responsible for the underlying event. Finance reviews and approves the overall numbers. The owner provides updates each week. This shared accountability structure prevents items from being forgotten or updated too late to affect the weekly forecast roll.

Data collection

Reliable data collection is what separates a forecast that tracks reality from one that gradually drifts away from it.

The main data sources used are:

• ERP modules — treasury, fixed assets, project accounting, and procurement. These contain details on approved capital spending, special contracts, and one-time payment obligations.
• Budget variance and exception reports — these surface unexpected items that may not yet have reached finance through the normal channel.
• Board minutes — acquisitions, major investments, legal cases, and strategic decisions often appear here with timelines and estimated amounts before they reach any formal register.
• CAPEX approval logs — planned asset purchases with approval status, expected amounts, and payment dates.
• Legal and contract registers — litigation logs, guarantee registers, and settlement tracking sheets shared regularly by the legal function.
• Balance sheet provisions and accruals — litigation provisions, warranty reserves, accrued liabilities, and long-term liabilities approaching settlement all signal future cash movements that may not yet be on anyone's radar.

Specific GL accounts are also monitored for one-off activity: legal expense accounts, restructuring provisions, settlement accounts, and miscellaneous income or expense lines. Unexpected movements in these accounts between forecast cycles are investigated immediately.

How they appear in the forecast

Each significant non-recurring event gets its own row in the forecast model. This makes individual items easy to track week by week and easy to explain when questions arise.

Smaller items of the same type are grouped into a single line when grouping does not hide anything material. For example, several small legal fees of similar nature can sit in one litigation line rather than cluttering the model with separate rows.

Descriptions are written to be self-explanatory without further explanation. A useful description includes the event type, counterparty or reference, and expected week — for example: Litigation settlement — ABC dispute — expected week 5, or Asset sale proceeds — equipment lot 3 — week 8.

Assumptions are documented either next to each line or in a dedicated notes section. The note records the basis for the estimated amount, the key timing driver, confidence level, and any important variables or contingencies. This keeps the forecast auditable and makes weekly updates straightforward.

Where a source document exists — a board approval reference, contract number, or email confirmation — it is referenced in the notes. This adds credibility without making the model unnecessarily heavy.

Monitoring

Non-recurring items are reviewed weekly before the forecast is rolled forward. The review checks whether timing, amounts, or probabilities need updating based on information received since the previous week.

Events requiring particularly close weekly attention are: large litigation settlements, asset sales or purchases nearing completion, major contract negotiations in the final stages, and any item whose expected payment week is within the next two to three weeks. These can change quickly and carry material impact.

The relevant department owner confirms the status of their items each week. Finance, legal, HR, procurement, and operations each contribute to this update. This verification (confirmation) improves accuracy.

Certain events trigger an immediate revision outside the normal weekly cycle: a signed agreement, a court ruling, a board approval, or a payment that arrives ahead of schedule. Unexpected events — emergency payments, sudden large receipts — are also incorporated as soon as they are known.

Events that carry high financial risk or could affect debt covenants are reported to senior management immediately. The reporting threshold is set at the start of each engagement based on the company's covenant levels and liquidity position.

Variance analysis

Variance analysis for non-recurring items looks at why actual outcomes differed from the forecast. The focus is on learning from each variance to improve the next forecast cycle.

Timing variances

The most common cause is delays in approvals, slower negotiations, or external factors such as regulatory review timelines. Where a particular type of event consistently slips, more buffer is built into the timing estimate for similar items in future forecasts.

Amount variances

These typically arise from final negotiation outcomes, additional costs that emerged during the process, or scope changes. The original estimate is compared to the actual figure, and the estimation method is adjusted if a systematic bias is identified.

Cancelled events

When an event is cancelled, the reason is documented. Common causes are deal failure, changed management priorities, or changed external circumstances. This prevents similar over-optimism in later forecasts.

Late discoveries

When a material event surfaces close to its payment date — rather than weeks earlier as it should — this points to a gap in the identification or data collection process. The specific gap is investigated and the process is adjusted.

Recurring mistakes in a particular area are addressed by updating assumptions, adding a check on a specific data source, or improving communication with the relevant department. The goal is not to assign blame but to make the next forecast more accurate.

Scenario analysis

Variance analysis looks backward — it explains what already happened. Scenario analysis looks forward — it shows different possible futures based on how uncertain events might play out. For non-recurring items, the two are closely connected: past deviations from similar events determine the assumptions used in scenarios.

Management compares scenarios by reviewing the weekly ending cash balance, the minimum cash level reached across the 13 weeks, and whether any week shows a potential funding gap. Presenting the three cases side by side makes the range of outcomes clear and supports decisions about whether additional liquidity headroom is needed.

Reporting

Non-recurring items are always shown in a dedicated section of the forecast report, separate from operating cash flows. This separation is maintained in every communication — whether the audience is management, the board, or a lender.

Management reporting

Management receives a full view of all material non-recurring items: descriptions, expected timing, amounts, confidence levels, and the effect on weekly cash balances. Any weeks where a non-recurring item creates a potential funding gap are highlighted explicitly.

Lender reporting

Lenders receive a focused summary that concentrates on items affecting liquidity and debt covenants. This includes all significant outflows, any asset sale proceeds material to liquidity, and clearly flagged risks to minimum cash levels or covenant compliance. The language is plain and factual.

Commentary accompanying a major non-recurring event covers four points: background, key assumptions, risks to timing or amount, and expected impact on cash. Keeping this short and factual is more useful than lengthy narrative.

Useful monitoring metrics for non-recurring items include: total net non-recurring cash flow per week, number of open items, forecast accuracy on completed items, and the share of total weekly cash movement attributable to non-recurring events.

Controls

Strong controls prevent events from being missed and keep the forecast reliable between weekly cycles.

The main control is a cross-department checklist completed at each weekly update. Department owners confirm the status of their items and flag any new events that have emerged. Short update meetings — focused only on non-recurring items — work well when the volume of open items is large.

Reconciliations provide an independent check. The forecast is compared to actual bank movements, open invoice reports, and balance sheet provisions each week. Unexplained differences prompt investigation before the updated forecast is finalised.

Monthly, a broader review is performed: board minutes from the previous month are read in full, legal and CAPEX registers are updated, and a variance analysis of completed items is documented. Standard assumptions are updated where recurring patterns in variances indicate the original estimates were consistently biased.

Supporting documentation is retained for all material items. This includes contracts, board approval extracts, legal correspondence, signed settlement agreements, and written confirmations from department owners. This documentation supports the forecast numbers during internal reviews and lender due diligence.

Risk

Some categories of non-recurring cash flow carry higher liquidity risk than others and deserve more attention throughout the forecasting process.

Largest liquidity risks — large litigation settlements, major asset purchases, business acquisitions, and significant tax assessments. These combine large amounts with uncertain timing and can create sudden, significant cash outflows with limited notice.

Events prone to amount uncertainty — legal and professional fees, environmental remediation, and employee restructuring. Final costs depend on negotiations, scope changes, or third-party decisions made late in the process. Conservative estimates are used, and the downside scenario includes an explicit buffer above the base case amount.

Events prone to timing uncertainty — contract settlements, lease terminations, asset sales, and business disposals. A signed agreement or court ruling may exist, but actual cash movement depends on counterparty action, regulatory steps, or completion conditions outside the company's control. Timing is treated as a range until cash is confirmed.

Events dependent on external decisions — litigation outcomes depend on court timelines, grant and tax refund receipts depend on government processing, and negotiated settlements depend on counterparty behaviour. These are tracked separately with scenario analysis applied from the point they enter the forecast.

Practical rules

These rules govern the lifecycle of a non-recurring item from the moment it is identified to the moment it is archived.

• Enter the forecast when there is reasonable evidence the event will happen — a management approval, a signed agreement, or a confirmed project plan. Rumours and early-stage discussions do not qualify unless the potential cash impact is very large.
• Remove from the forecast when the event is confirmed as cancelled, completed, or no longer expected within the 13-week window. Removed items are kept in a separate log for reference.
• Update timing when new milestone information is received — a revised completion date, a delay from a counterparty, or earlier-than-expected progress. Timing updates happen during the weekly cycle, not ad hoc.
• Revise the amount when better estimates are available from completed negotiations, updated quotes, or final contract terms. The reason for each revision is documented.
• Split into multiple payments when the contract or plan shows staged payments — milestone-based costs, phased project payments, or instalment arrangements.
• Require supporting documentation for all material items before they are included in the main base case forecast. Oral updates alone are insufficient for items above a materiality threshold agreed at the start of the engagement.
• Require management approval for large or high-risk items that could affect covenants or create a funding gap. Approval is confirmed before the item enters the main forecast.
• Communicate uncertainty with explicit confidence levels — high, medium, or low — alongside a brief note on the main risk to timing or amount. This keeps expectations realistic and prevents surprises.
• Report to lenders with plain, factual language. Each material item includes a description, expected cash impact, effect on covenants if relevant, and risk level.
• Archive completed items in a separate log or in previous forecast versions. The archive records the final actual amount, the actual payment date, and the explanation of any variance from the forecast. This is the primary source for improving estimates on similar future events.

Frequently asked questions