Executive Summary

Traditional static cash flow forecasting methods, while useful, struggle to keep pace with shifting market conditions, changing customer behaviour, and evolving operational demands. This paper explores a more advanced approach – combining cyclical analysis with rolling forecasts – to give businesses the agility to respond quickly and the visibility to make informed strategic decisions.

With the help of smart technology, this integrated approach creates a robust framework that can capture predictable seasonal patterns while adjusting in real time to unexpected events. Companies using these methods report stronger cash visibility, lower financing costs, and improved capacity to act on growth opportunities.

Introduction

Cash flow forecasting is a perennial headache – even for the most sophisticated treasury teams. According to PwC’s Global Treasury Survey 2025, 38% of companies with revenues over $10 billion and 52% of those between $1 billion and $10 billion are still manually collecting and consolidating forecast data^*1. The survey also shows that treasurers are struggling with poor data quality (76%), ineffective tools (53%) and a lack of incentives for business units to contribute (46%) – all of which are still holding back forecasting performance across the board.

But the prize for getting cash flow forecasting right, or as accurate as possible, is not to be underestimated. Better cash forecasting can help to:

• Free up working capital 


• Sharpen debt issuance decisions 


• Strengthen investor guidance and improves risk management

Many leading treasury teams are looking to achieve these goals by embedding automation, analytics and accountability into their forecasting processes, thereby positioning themselves as trusted advisers to the wider business.

Arguably, this matters more than ever with interest rates still volatile and capital harder to come by, since cash efficiency is extremely high on the agenda. And companies that can see, move and optimise cash in real time will be best placed to respond quickly and with confidence.

One way to help achieve this is by moving beyond static budgets towards forecasting methods such as cyclical analysis and rolling forecasts that are built for both accuracy and agility.

Understanding Cash Flow Cycles

The Nature of Business Cycles

Every company has a cash flow rhythm, which is the pattern of money moving in and out. Typically, these cycles are shaped by factors such as industry dynamics, seasonality, customer payment behaviours and operational requirements. For treasury teams, a clear view of these patterns is essential for effective cash management.

Thankfully, most businesses experience at least some predictable patterns in their cash flows. The retail sector, for example, sees inflows peak in the run-up to holiday seasons, while construction activity often follows weather conditions, and many B2B service providers operate on quarterly payment cycles tied to their clients’ budget processes.

But real life is not so simple! On top of these predictable patterns are irregular events such as large contract wins, equipment purchases, or market disruptions that can significantly impact cash flow timing.

The task for treasury is to map these cycles with precision, identify where volatility may occur and plan for both the predictable and the unforeseen.

Components of Cash Flow Cycles

Cash flow isn’t a single stream. Rather, it moves through distinct channels that reflect different aspects of the organisation’s activity. Understanding each type of cycle and how they interact provides can help provide treasury with a clearer picture of liquidity needs, funding requirements and investment capacity:

• Operating cash flow cycles cover the day-to-day movement of cash through business operations – from collecting customer payments to paying suppliers, running payroll and covering regular operating expenses. The length and predictability of these cycles vary widely by industry and business model. 


• Investment cash flow cycles relate to capital expenditure, acquisitions and other strategic investments. These cycles are generally longer and less predictable than operating cycles, but they are critical to growth and competitiveness. 


• Financing cash flow cycles capture cash movements between the organisation and its lenders, investors and shareholders. This includes servicing debt, distributing equity returns and raising new capital. Understanding these cycles is key to maintaining the right capital structure and meeting any covenant obligations.

Table 1: Cash flow cycle components at-a-glance

Category	Operating	Investing	Financing
Definition	Cash generated and used in normal business activity such as producing and selling goods or services.	Cash movements linked to acquiring or disposing of long-term assets or investments.	Cash flows between the organisation and its lenders, investors or shareholders.
Typical inflows	Customer receipts, service fees, royalties, other operating income.	Proceeds from asset sales, interest received, dividends from investments.	Proceeds from loans, bond issues, share issues.
Typical outflows	Payments to suppliers, payroll, tax payments, rent, other operating expenses.	Capital expenditure, property purchases, investments in other businesses.	Loan repayments, interest, dividend payments, share buybacks.

Spotting Cyclical Patterns

Identifying cycles starts with analysing historical cash flow data over multiple periods. In practice, this means looking at different time horizons – from daily patterns within a single month, to seasonal peaks and troughs over several years – and understanding how they interact. Techniques such as seasonal decomposition, trend analysis and correlation studies can reveal patterns that are not always immediately obvious, at least to the human eye.

Some treasury teams are taking this further, layering in operational and market data to explain the ‘why’ behind the patterns. For example, aligning payment trends with promotional calendars, production cycles or commodity price movements can make forecasts more accurate and actionable.

Of course, this does not need to be done manually – that is the good news! Technology can assist in a variety of ways, as we will explore further in this paper. And a handful of sophisticated treasury teams are now testing artificial intelligence (AI) machine learning (ML) models to uncover subtler relationships in cash flow data like the combined effect of sales promotions, weather events and customer payment behaviour, for instance.

In a nutshell, these tools aim to flag leading indicators of change before they appear in the numbers. And, from conversations with leading treasury teams, early results appear promising. That said, it is important to remember that the technology is still maturing – and areas such as data quality, model transparency and validation must be robust before the ML can deliver consistently reliable results.

Rolling Forecast Methodology

Conceptual Framework

As the name suggests, rolling forecasts replace the rigidity of fixed-period budgets with a dynamic, continuously updated view of the future. Instead of locking into a target that begins to age the moment it’s signed off, the forecast horizon shifts forward with each update, always offering the most current picture of the road ahead.

In practice, the concept itself is simple: markets evolve, priorities shift, and unforeseen events occur. A rolling approach allows companies to adjust course quickly, seize opportunities as they emerge, and address risks before they escalate, while still keeping sight of long-term objectives.

Instead of static annual budgets, the rolling method fosters a continuous cycle of refinement. By regularly updating projections and extending the horizon, rolling forecasts ensure decisions are based on the latest information. In a world where business conditions change constantly, planning processes must adapt accordingly. A consistent, forward-looking horizon equips companies to respond faster, plan smarter, and remain strategically focused.

Implementation Structure

Getting the most out of a rolling forecast depends on a few key design choices such as how far ahead to look, how often to refresh the numbers, and how much detail to include. The right combination will vary by organisation, but each decision has a direct impact on how useful and actionable the forecast becomes.

• Forecast horizon selection represents a critical decision in rolling forecast implementation. Most companies adopt 12 to 18-month rolling periods (although there are some exceptions – see table 2), balancing the need for forward visibility with forecast accuracy limitations. Shorter horizons may not provide sufficient time for strategic planning, while longer horizons often lack the accuracy needed for operational decisions. 


• Update frequency depends on business characteristics and volatility. Some treasury teams refresh forecasts monthly, others quarterly. High-volatility environments or periods of rapid change often call for more frequent updates, while stable operations can work effectively with less frequent revisions. 


• Granularity levels should match the company’s decision-making needs. Near-term periods often call for detailed, line-item forecasts, while longer-term projections can focus on broader categories and strategic priorities. The aim is to provide enough detail to support decisions without creating unnecessary administrative burden.

Table 2: Typical rolling forecast horizons by business type

Business type / environment	Common horizon	Rationale
High-volatility sectors (commodities, early-stage tech, start-ups)	6-9 months	Market conditions and revenue streams shift quickly, making longer forecasts unreliable.
Most mid-sized and large corporates	12-18 months	Balances forward visibility for strategic planning with acceptable forecast accuracy.
Capital-intensive or long-cycle industries (infrastructure, energy, aerospace)	18-24 months	Long project lead times and relatively predictable revenue patterns support extended horizons.

Integration with Business Planning

Rolling forecasts typically work best when they are embedded in the company’s wider planning cycle, not run in isolation by finance. Linking forecasts to operational plans, strategic priorities and performance targets can help treasury to ensure that cash projections are grounded in reality and support decision-making across the business.

Rather than just a systems issue, effective integration depends on active communication between finance, operations, sales, procurement and strategy teams. This cross-functional input helps align forecasts with sales expectations, production schedules, capital investment plans and other initiatives – reducing the risk of conflicting assumptions or missed interdependencies.

Clear governance is equally important here. Assigning ownership for inputs, updates and variance explanations should help to the process on track and builds accountability for forecast accuracy.

Common Integration Pitfalls to Avoid

• Siloed ownership - forecasts built by finance alone without operational input. 
• Out-of-sync updates - sales, production or capex plans updated on a different timetable to forecasts. 
• Conflicting assumptions - revenue or cost projections that don’t match across departments. 
• Lack of accountability - no clear owner for forecast accuracy or variance explanations. 
• Overcomplication - excessive detail that slows updates and limits agility.

Combining Cycles and Rolling Forecasts

Synergistic Benefits

Bringing cyclical analysis and rolling forecasts together can help to create a forecasting framework that blends predictability with agility. Cyclical analysis establishes the baseline by mapping normal business rhythms, while rolling forecasts make it possible to adjust quickly when conditions change (precisely what’s needed in today’s environment).

In turn, this combined approach enables treasury teams to build forecasts on proven historical patterns, then refine them in real time as new data emerges. Done well, this results in forecast that is both statistically robust and responsive to current realities – improving accuracy, enhancing liquidity control and enabling faster, better-informed decisions.

Methodology Integration

What makes this combined approach powerful is how the two methods are woven together. Each element builds on the other in a virtuous circle, creating a process that is both grounded in historical reality and ready to respond to change.

• Baseline cycle modelling - builds mathematical models that capture historical cyclical patterns to form the foundation for rolling forecasts. These structured starting points reflect established business rhythms. More advanced teams may use multiple models to capture different aspects of their operations, from seasonal sales peaks to broader economic cycles. 


• Dynamic adjustment mechanisms - introduces processes that keep forecasts aligned with current conditions without losing sight of cyclical patterns. Examples include variance analysis to flag deviations, trigger points for forecast updates, and escalation procedures to secure management attention when major shifts occur. 


• Scenario planning integration. - combines cyclical insight with rolling forecasts to model best-case, worst-case and most-likely outcomes. Running these scenarios under different assumptions about patterns and external factors equips decision-makers with a range of potential futures and a clearer view of key risks.

Technology and Tools

While many treasury teams still rely on Excel for forecasting, modern technology is now available (and accessible to many) that can manage large datasets, process complex calculations, and deliver real-time updates.

As such, today’s treasury management systems (TMS) go far beyond spreadsheet replacement. Instead, they provide secure, centralised platforms for collaboration across departments and geographies, with built-in version control and audit trails.

Integration is critical for these systems to achieve their maximum results, however. For instance, linking the TMS with enterprise resource planning (ERP) systems brings in live operational data, while application programming interfaces (APIs) connect forecasting tools to banking systems, customer relationship management (CRM) platforms, and other core business applications. In turn, this connectivity enables a single, consolidated view of cash that reflects the latest sales, procurement, and payment information.

As mentioned earlier, maturing capabilities such as ML and predictive analytics are beginning to play a larger role here too, identifying patterns in historical and external data that might otherwise be missed. Cloud-based platforms are also enabling faster deployment, easier updates, and scalable computing power for complex models.

And, in Salmon’s experience of working with treasury leaders, the most effective teams use technology not only to automate and integrate but to elevate the quality of discussion around cash.

Best Practices and Implementation

Organisational Readiness

Implementing advanced cash flow forecasting is an organisational initiative, not just a treasury responsibility. Success therefore depends on senior leadership setting the tone, allocating resources, and signalling that accurate, timely forecasting is a shared priority. Without this visible sponsorship, it can be difficult to get consistent engagement from other departments.

Transitioning from static budgets to rolling forecasts also means changing habits and mindsets. It can require a number of new approaches, including:

• Process change – new workflows for collecting, validating and integrating data. 


• Behavioural change – an understanding that forecasts are living documents, not annual reports to be filed away. 


• Cultural change – seeing forecasting as part of the company’s collective intelligence, not just a finance function responsibility.

Training also plays a pivotal role here and its value should not be overlooked. Treasury and finance teams need technical skills to operate new systems and analytical tools, for instance. While operational teams need to understand how their inputs influence the forecast and why their accuracy matters. Explaining the ‘why’ behind the process can help increases buy-in and improve data quality, meaning that softer skills like communication and change management are more important than ever.

Data Quality and Governance

Even the most sophisticated forecasting model will likely fail if it is fed with inconsistent or incomplete data. Strong data governance is essential, with defined processes for ensuring accuracy, completeness and timeliness. This should include:

• Clear ownership of each dataset. 


• Regular validation and exception handling. 


• Agreed standards for data formats, naming conventions and classifications.

As a result, master data management is also critical when multiple systems are in play. Consistency is the key and needs to be maintained across customer codes, product categories and account structures to enable accurate aggregation and analysis across ERP, TMS, CRM and other data sources. Poor master data discipline often causes avoidable reconciliation issues and undermines confidence in the forecast.

Performance Measurement

As every treasurer knows, tracking forecast accuracy is essential – but accuracy alone is not the whole story. A forecast that hits its numerical target yet fails to influence business decisions is arguably little more than an academic exercise. Seasoned treasurers know that the real test is whether the forecast helps the treasury team act earlier and with greater confidence.

Key performance indicators might include:

• Forecast accuracy percentages by time horizon (e.g. one week, one month, one quarter ahead) to measure reliability over different planning windows. 


• Variance analysis by magnitude and direction to identify whether deviations are minor fluctuations or significant misjudgements – and whether the bias is towards over- or underestimating. 


• Trend analysis to detect consistent patterns of error and highlight where models, data sources or assumptions need review. 


• Frequency and drivers of forecast revisions, showing whether changes are the result of predictable seasonal effects, one-off shocks, or a flaw in the underlying methodology.

While these metrics provide the raw performance picture, they must be viewed in context to give the full flavour of what is happening in the business. For example, a forecast that is 95% accurate yet fails to warn of a looming cash crunch has missed its purpose. Equally, a forecast that is less precise numerically but triggers timely action to secure funding or adjust investment priorities may be delivering far greater strategic value.

Leading companies tend to measure both the statistical quality of the forecast and its impact on decision-making. They will typically ask questions such as:

• Does the forecast provide early warning of cash shortfalls? 


• Is it guiding capital allocation and investment timing? 


• Has it improved the organisation’s ability to negotiate better financing terms? 


• Are operational teams using it to plan inventory, production, or procurement more effectively?

If the answer to these questions is ‘no’, then accuracy becomes nothing more than a hollow metric. In other words, the forecast is not fulfilling its role as a decision-support tool. As such, the ultimate goal is to create a forecast that is not just a reflection of what might happen, but a catalyst for making the right moves before events unfold.

Continuous Improvement

Like many good business habits, advanced forecasting is a discipline that strengthens with consistent use, feedback, and refinement. It’s not a one-off project but an ongoing capability that matures as the company learns from both its successes and missteps.

Disciplined treasury teams tend to build review and improvement into the rhythm of their forecasting process. This might include:

• Reviewing performance regularly – not just to tick a compliance box, but to understand what drove variances and whether those drivers are likely to recur. 


• Revisiting cyclical assumptions – checking if established patterns still hold true or if shifts in markets, customer behaviour, supply chains or regulation mean the cycle has fundamentally changed. 


• Introducing new data sources or analytical techniques – whether that’s integrating sales pipeline data, economic indicators, or using advanced analytics to uncover relationships that were previously hidden. 


• Using post-event reviews – examining where the forecast proved reliable and where it fell short, and using those lessons to fine-tune both models and processes.

What underpins this approach is a commitment to continuous evolution. In other words, a forecast is never ‘finished’ – it should flex and adapt as quickly as the organisation and the environment around it. External shocks, internal strategy shifts, and emerging opportunities all create conditions that demand fresh assumptions, updated inputs, and sometimes a rethink of the methodology itself.

By treating forecasting as a living system, fed by accurate data, informed by real-world experience, and open to innovation, treasury teams can improve accuracy over time and help increase the level of trust that decision-makers place in the forecast.

Table 3: Forecasting maturity model

Stage	Characteristics	Risks	Opportunities
Basic	Forecasts built manually, often in Excel; limited departmental input; short time horizons; updates infrequent.	High error rates, poor visibility, reactive decision-making.	Introduce structured processes, basic automation, and cross-functional involvement.
Developing	Forecasts generated from integrated ERP/TMS data; cross-functional input; some rolling forecasts; variance tracking in place.	Inconsistent data quality, process gaps, limited scenario capability.	Strengthen data governance, expand horizons, integrate scenario analysis.
Advanced	Fully integrated rolling forecasts combining cyclical analysis, real-time data feeds, and scenario modelling; embedded in strategic and operational planning.	Complexity may slow adoption if change management is weak.	Leverage predictive analytics, machine learning, and continuous improvement for competitive advantage.

Checklist for building forecasting maturity 

• Leadership actively sponsors and communicates the initiative 
• Clear ownership of inputs, updates and variance explanations 
• Staff trained on both tools and the purpose behind the process 
• Data governance in place – accuracy, completeness, timeliness 
• Master data aligned across systems 
• KPIs tracked and acted on, not just reported 
• Regular review cycles and model refinements 
• Forecasting seen as a dynamic, business-wide process

Risk Management and Contingency Planning

Identifying Cash Flow Risks

As outlined, effective cash flow forecasting is not just about predicting what will happen, it’s also about anticipating what could happen and being ready to respond. That means building a comprehensive risk assessment into the forecasting process to flag potential disruptions to normal cash patterns.

Risks can stem from many sources including customer credit issues, supplier delays, regulatory shifts, market volatility, operational breakdowns, geopolitical events, extreme weather, and more. Some are predictable and cyclical while others arrive without warning – we all know the concept of ‘black swans’.

A strong risk identification process combines quantitative analysis of historical variations (e.g. how cash flows behaved during past market shocks) with qualitative assessment of emerging threats. Pairing these insights with scenario modelling allows companies to estimate the potential impact of each risk and plan the best course of action in advance.

Building Contingency Plans

Contingency planning extends beyond simple cash flow forecasting to include specific action plans for various scenarios. Depending on the situation, this could mean activating pre-arranged credit lines to secure immediate liquidity, freezing or deferring non-essential capital projects to preserve cash, or implementing targeted expense reductions in specific business areas. In other cases, it might involve accelerating the collection of receivables, renegotiating payment terms with customers or suppliers, or reallocating cash within the group to support entities under strain.

Often, the most effective contingency plans aren’t static – instead, they’re integrated with the rolling forecast process so they can be revisited and updated regularly. With this approach, each forecast update becomes an opportunity to test whether the planned responses are still appropriate in light of the latest operational realities, market conditions and strategic priorities.

Stress Testing and Scenario Analysis

As the name suggests, stress testing involves pushing forecasts to their limits to see how they perform under adverse conditions. Essentially, it is a deliberate exercise in imagining the extremes such as a sudden economic downturn, the loss of a major customer, a significant supply chain disruption, or perhaps a regulatory change that tightens liquidity or alters capital requirements. Running these scenarios helps to uncover vulnerabilities in the organisation’s cash position and assess whether existing contingency measures are strong enough to cope.

Scenario analysis can then be used to build on this by broadening the lens beyond purely negative events. This helps treasurers to model the impact of positive developments such as an unexpected surge in demand, entry into a new market, or the opportunity to acquire another business. And by understanding the cash flow implications of both upside and downside scenarios, treasury can take a more balanced and proactive approach to planning – positioning themselves to respond quickly to whatever the future might hold.

Technology and Innovation

Integration and Automation

Unlike spreadsheets, modern forecasting solutions are designed to connect seamlessly with existing business systems and remove manual effort from routine processes. Integration with the ERP, banking, CRM and sales systems ensures forecasts are built on the most current operational data, while APIs allow additional sources to be connected as business needs evolve.

Automation plays an equally important role here – and one that is growing rapidly, too. For instance, robotic process automation (RPA) can take on repetitive tasks such as data collection, validation and regular forecast updates, freeing finance and treasury teams to focus on analysis, insight generation and strategic decision-making. Meanwhile, workflow automation can also streamline approvals, variance reviews and data reconciliation, speeding up the entire forecasting cycle.

In addition, cloud-based solutions bring even further benefits, including real-time collaboration between geographically dispersed teams, scalability for handling large and complex models, and easier integration with external data feeds such as economic indicators, market research, and industry benchmarks.

All-in-all, by creating a single, connected forecasting environment, these technologies can help treasury teams to ensure the numbers are both current and trusted.

Future Developments

While technology is constantly evolving and adapting, the next stage in cash flow forecasting will likely be defined by greater automation, sharper analytics, and even closer alignment with strategic planning – all achieved in much easier and more accessible ways. Real-time forecasting capabilities will also enable treasury teams to model the impact of new information as it emerges, whether it’s a sudden change in sales orders, a currency movement, or an unexpected supplier delay, and – crucially – equip them to respond immediately.

Interestingly, further advances in predictive analytics and ML will also enable forecasts to incorporate a broader range of data, from macroeconomic indicators to sector-specific signals and even less traditional sources such as social media sentiment. The aim will be to move from reactive to anticipatory forecasting, spotting trends and risks before they fully materialise.

Industry examples from the cutting-edge suggest that companies embracing these developments will be able to operate with a clearer view of both risk and opportunity. Forecasting will then naturally shift from a periodic reporting exercise to a continuous, decision-driving capability that adapts as quickly as the environment it is built to navigate.

Conclusion

Cash flow forecasting that combines cyclical analysis with rolling forecasts marks a step-change from traditional budgeting. As we have explored, companies using these methods in the right way can benefit from sharper visibility into future cash positions, greater agility in responding to shifting conditions, and stronger support for strategic decision-making.

While it might sound complex, the foundations of success are clear: committed leadership, high-quality data, fit-for-purpose technology, and a culture of continuous improvement. Although implementation inevitably takes effort and resources, the potential pay-off is significant – ranging from improved cash management to reduced financial risk and the strategic flexibility to act quickly when opportunities or threats emerge.

And, importantly, advanced forecasting can set treasury teams apart – enabling faster, better-informed decisions than relying on static budgets and arguably delivering a competitive edge. After all, the ability to anticipate and adapt to cash flow changes supports more ambitious growth plans, tighter risk control, and stronger financial performance.

Of course, forecasting improvement is not a one-off project but an ongoing journey. New technologies and analytical techniques are emerging all the time, each with the potential to enhance accuracy and speed while deepening insight. But treasury teams that treat forecasting as an evolving discipline, tightly linked to strategic goals, will be better equipped to respond decisively as conditions change – an ability that cannot be underestimated in these volatile and unpredictable times.

Ready to strengthen your forecasting?

Salmon Software helps treasury teams move beyond static budgets with integrated tools for rolling forecasts, variance analysis, and real-time cash visibility. Whether the goal is sharper liquidity control, faster decision-making, or greater confidence in strategic planning, Salmon’s platform brings the data, automation, and insight together in one place.

Get in touch to explore how Salmon can support your forecasting journey.