Strengthening financial forecasting through better use of data and systems

In an increasingly volatile business environment, organisations rely on forecasting to support decision-making, manage risk and respond to change. Yet despite advances in planning tools, analytics and Artificial Intelligence (AI), many organisations continue to struggle with forecast accuracy and confidence in their projections.

A common aspiration is to develop the perfect forecasting model. In reality, every organisation has different drivers, operating environments and information needs, making forecasting highly contextual.

More importantly, even the most sophisticated forecasting methodology is only as reliable as the data behind it. Incomplete, inconsistent or poorly governed information will inevitably undermine forecast quality. As organisations become more data-driven and increasingly adopt advanced analytics and AI-enabled forecasting capabilities, forecasting has evolved beyond a traditional finance exercise. It is now a business capability that depends on quality data, integrated systems, effective governance and close collaboration between finance and digital teams.

Why good forecasts go wrong

Most forecasting challenges can be traced back to recurring issues, and very few relate to the mathematics behind the model. Common challenges include:

Data quality: Different reports may contain different versions of the same information. When data lacks consistency, the forecast becomes difficult to trust.
Fragmentation: Data is often spread across multiple systems. Sales data, headcount information and working capital figures are frequently maintained in separate platforms. Combining this information manually is time-consuming and increases the likelihood of errors.
Static thinking: Many organisations still treat forecasting as an annual exercise, leaving budgets unchanged even when conditions shift. As markets move faster, organisations increasingly require rolling forecasts that can be refreshed regularly.
Heavy reliance on spreadsheets: Excel remains essential in finance, but when forecasts depend on complex workbooks maintained by a small number of individuals, organisations face operational and governance risks.

Data first, then models

Improving forecasting starts with improving the data and systems that support it.

In many organisations, information is distributed across separate databases, operational platforms and reporting tools that do not communicate effectively with one another. This creates fragmented reporting, multiple versions of the truth and significant effort spent reconciling information rather than analysing it. At the same time, data is often collected through multiple processes, systems and owners without consistent standards, governance or business definitions, resulting in organisations that are data rich but insight poor.

A strong forecasting capability requires more than accurate data. It requires a well-governed and integrated data environment that enables information to be trusted, shared and reused across the organisation. This includes agreed definitions for key metrics, clear ownership of critical datasets, data quality standards, validation controls and accountability for maintaining information throughout its lifecycle.

From a digital perspective, organisations often face additional complexity due to the number of applications supporting different business functions. Financial, operational, customer and workforce data may reside across ERP systems, CRM platforms, operational applications, spreadsheets and external data sources. Without effective integration, forecasting processes frequently become dependent on manual extraction, reconciliation and consolidation activities, increasing both effort and the risk of errors.

Establishing a single source of truth requires more than consolidating information into a reporting tool. It requires a deliberate approach to data architecture, integration and governance. By connecting systems, standardising data definitions and improving information flows, organisations can reduce reconciliation effort, improve transparency and increase confidence in forecast outputs.

Technology plays a critical role in enabling this foundation. Modern data platforms, integration technologies and planning solutions can automate the movement of information between systems, provide near real-time visibility into business performance and create a scalable foundation for forecasting and decision making. These capabilities help organisations move away from static and spreadsheet-driven processes towards more dynamic and continuously updated forecasting environments.

However, technology alone is not enough. Poor quality data processed through sophisticated systems will simply produce unreliable forecasts faster. Effective governance remains essential. Well-designed systems support governance through automated controls, validation rules, workflow approvals, data lineage tracking and auditability, helping organisations maintain consistency and trust in their information over time.

Increasingly, organisations are recognising that forecasting maturity is closely linked to data maturity. The growing accessibility of AI, machine learning and advanced analytics means that developing sophisticated forecasting models is no longer the primary challenge. The greater challenge lies in establishing the data foundations that enable these technologies to deliver reliable and explainable results.

AI-enabled forecasting solutions are only as effective as the data they consume. Accurate, complete, timely and well-governed data remains critical to producing meaningful outputs. Organisations that invest in data quality, governance and integration are not only improving forecasting accuracy but also creating a foundation for broader analytics, automation and AI initiatives.

Timeliness is equally important. Automated data pipelines and system integrations help ensure forecasts reflect current business conditions, allowing organisations to respond more quickly to emerging risks and opportunities. As forecasting cycles become increasingly continuous rather than periodic, access to trusted and up-to-date information becomes a strategic advantage.

Once a strong data foundation exists, forecasting models become significantly more effective. The conversation can then shift from collecting and reconciling information to generating insight, evaluating scenarios and supporting better decision making.

A particular consideration for Malta

These challenges are universal but are often particularly visible among small and medium-sized businesses in Malta. Information may be concentrated among business owners and managed through spreadsheets, paper records and separate accounting applications. While these approaches may be sufficient initially, they can become constraints as organisations grow.

Without appropriate systems and governance structures, producing a forecast can require substantial manual effort. Information must be gathered, reconciled, validated and interpreted before meaningful analysis can begin. This can lead to unreliable forecasts and reduced confidence in decision-making.

The good news is that access to technology has improved significantly. For many organisations, the challenge is increasingly less about technology availability and more about data quality, governance, process discipline and effective adoption.

Building smarter forecasting models

Modern forecasting should move beyond simply taking last year’s numbers and applying a percentage increase. Instead, organisations should seek to understand the underlying drivers of business performance. Driver-based forecasting focuses on the operational factors that influence results, such as sales volumes, customer demand, utilisation rates, inflation assumptions and industry-specific metrics.

Digital capabilities are increasingly enhancing driver-based forecasting through advanced analytics and predictive modelling techniques. By analysing larger volumes of operational, customer and market data, organisations can identify emerging trends and leading indicators that may influence future performance and improve the quality of forecasting assumptions.

The quality of external data used to inform assumptions is equally important. Advances in data integration technologies allow organisations to combine internal financial and operational information with external economic, industry and market data sources. This provides a broader view of the factors influencing performance and enables forecasting models to respond more dynamically to changing business conditions.

Scenario planning remains critical. Rather than relying on a single forecast, organisations should evaluate multiple possibilities through best-case, base case and worst-case scenarios. This helps management prepare for uncertainty and understand potential risks before they materialise.

As forecasting capabilities mature, organisations are increasingly exploring AI-enabled forecasting and advanced analytics. AI can help identify trends, detect anomalies, generate forecasts and evaluate scenarios at a scale that would be difficult to achieve manually. However, these technologies should enhance rather than replace human judgement. Effective forecasting still depends on business context, professional insight and accountability for decisions.

Forecasting models should also be reviewed periodically to assess their effectiveness. Monitoring forecast accuracy, comparing outcomes against actual performance and refining assumptions over time helps ensure that models remain aligned with changing business realities.

Automated data flows, workflow orchestration and integrated planning platforms allow forecasts to evolve alongside business conditions. As organisations move towards continuous planning models, forecasting becomes less of a periodic reporting exercise and more of an ongoing decision support capability powered by timely data and analytics.

Ultimately, smarter forecasting models are not defined solely by their analytical sophistication. Their effectiveness depends on the combination of quality data, robust governance, integrated technology platforms and sound business judgement.

Where finance and digital need to meet

Effective forecasting requires close collaboration between finance and digital teams.

Historically, forecasting has often been viewed as a finance process supported by technology. Increasingly, however, forecasting is becoming a cross-functional capability that depends on the combined strengths of finance, data and digital disciplines.

Finance teams bring an understanding of business performance, commercial drivers, regulatory requirements and management’s information needs. Digital teams contribute expertise in data governance, systems integration, automation, analytics and emerging technologies. Together, they establish the foundations required to generate forecasts that are timely, reliable and actionable.

The challenge for many organisations is not a lack of data, but an inability to connect, govern and use it effectively. Close collaboration between finance and digital functions is critical to establishing an integrated data environment that supports informed decision-making.

As organisations increasingly adopt AI-enabled forecasting and advanced analytics capabilities, this partnership becomes even more important. Finance and digital teams must work together to ensure that AI solutions operate on trusted data and deliver outputs that are explainable, transparent and aligned with business objectives.

In practice, finance and digital teams should jointly define data standards, establish ownership and accountability for critical information, design planning processes, automate data flows and implement controls that maintain data quality over time. Forecasting should be supported by a governance framework that ensures consistency, auditability and confidence in decision-making.

Leading organisations are increasingly treating forecasting as part of a broader enterprise data strategy rather than as a standalone finance activity. By aligning financial planning with data management, technology architecture and analytics capabilities, organisations can build a sustainable forecasting capability that evolves alongside the business.

Where to start

Organisations do not need a large-scale transformation programme to begin improving forecasting. A few practical steps can make a significant difference:

Establish a single source of truth and agree clear definitions for key metrics.
Define ownership and accountability for critical data.
Improve data quality before investing in new forecasting tools.
Assess opportunities to simplify and integrate systems.
Automate manual and repetitive data collection, validation and consolidation activities.
Introduce rolling forecasts and regular scenario planning.
Use driver-based models rather than relying solely on historical trends.
Build forecasting processes with governance, controls and auditability in mind.
Involve finance, data and digital specialists early in the process.
Prepare data foundations for advanced analytics, and AI-enabled forecasting.
Use technology to handle forecasting mechanics while finance focuses on insight and decision making.

Conclusion

Forecasting will never predict the future with certainty, nor should it. Its purpose is to help organisations understand potential outcomes, prepare for uncertainty and make better informed decisions.

While forecasting models and analytical techniques remain important, their effectiveness is ultimately determined by the quality of the underlying data, the governance supporting it and the systems through which it flows. As advanced analytics, automation and AI become more accessible, forecasting maturity is increasingly linked to data maturity.

Success therefore depends on more than finance expertise or technology investment alone. It requires collaboration between finance, data and digital teams to establish trusted data foundations, effective governance and integrated technology ecosystems.

Organisations that invest in these capabilities will be better positioned to improve forecasting accuracy, strengthen decision-making and respond to change with greater confidence and agility.

RSM Malta supports organisations in strengthening their forecasting capabilities by bringing together financial expertise, data, systems and digital advisory. To discuss how your organisation can improve the reliability and effectiveness of its forecasting processes, contact RSM Malta.

Insights

A forecast is only as good as the data behind it