September 08, 2023

From Data to Decisions: The Complex Art of Long-Term Planning

Four days before France and New Zealand were due to take to the field in Paris to kick off the 2023 Rugby World Cup, a BBC article reported that Stats Perform had produced its Opta Data, AI-driven Tournament Predictor showing the percentage chance of each nation winning the Cup – link at the end if you’re interested! It takes into account thousands of data points including recent form, past success, pools, fixture timings, difficulty, and much, much more. In the age of AI, this accounts for plenty of information to go on to make some credible predictions.

Now imagine also trying to predict today, to the same degree of accuracy, the outcome of the 2027 World Cup. Suddenly data analysis turns to guesswork and probability as the model tries to predict how each team’s form and performance will pan out over the coming years – and what other micro or macro factors might come along to impact the game of rugby and specific fixture results.

This is the challenge businesses, and in particular their finance teams, face every day when it comes to planning for the longer-term future.

Long-range business forecasting and planning is a critical feature of strategy design, effective business management and change navigation. Without it, businesses might fail to adequately resource themselves to execute their strategy, they could miss the opportunity to gain competitive advantage, or they may falter due to a lack of resilience and contingency planning. It is a process that requires agility, a proportionate approach, and an accurate quantification of risk – three features that have arguably never been harder to achieve than today.

One of the most challenging responsibilities held by the Finance function, long-range forecasting is of course based on the notion that market, industry, and economic trends can be predicted – at least within a sensible range of assumptions – several years in advance. Forecasts must also be updated frequently and remain flexible to reflect real-time events.
At a time of increasing globalisation and digitalisation, and with constant volatility, uncertainty, complexity, and ambiguity (also referred to as VUCA), the challenge has become even greater.

Some of the most common difficulties in forecasting and planning that are undermining successful decision-making today include:

  • Supply chain pace and security – emerging due to Covid-related backlogs, these have been exacerbated by Russia’s invasion of Ukraine and its ongoing impact on supply and logistics
  • Data and device vulnerabilities – ever-increasing and evolving cyber attacks that threaten business continuity and reputation
  • ESG – customers rightly demanding higher standards and more transparency in their approach to the environment, social responsibility, and governance – with reputation and demand greatly enhanced through doing the right thing but quickly threatened when things go wrong
  • Regulatory change and complexity – increasing demands from Regulators across many industries. Changing rules and the costs of compliance are difficult to predict and may constrain growth and innovation strategies
  • Talent shortage and trends – as the world continues to recover from the pandemic, there remain significant talent gaps and challenges across a number of industries. These are driven by various factors, including hybrid working and reduced appetite for office working or relocation, the demand for new/emerging skills in short supply, and the impact of Brexit – with these drivers culminating in higher and less predictable people costs
  • Digital transformation and disruption – continually evolving and emerging technologies mean that companies need to constantly assess their business models and horizon-scan for new competitive threats
  • Systems issues – many finance teams find their mission to produce quality forecasts constrained or damaged by systems issues – which inevitably saps time and causes frustration. The most common challenges include over-reliance on Excel with a higher risk of human error, failed implementation of a new system, and ongoing use of systems that have failed to keep pace with the growth or development of the business

What does this landscape mean for businesses and how do they go forward with forecasting and planning?

Whilst the situation may seem daunting, there are a variety of tools and approaches businesses and their Finance teams might utilise as they look to navigate their forecasting and planning:

Improve data quality – Many organisations are still working with data that is either incomplete, badly interrogated, or simply inaccurate. But without high-quality, accurate, and relevant, value-add data, no matter how sophisticated and well-structured your forecasting model may be, or how many scenarios you run through it, your projections will be unreliable, redundant, or worst of all, misleading for decision-making. Focusing on data quality over quantity is key and Finance teams should insist on this as a starting point.

Utilising Key Performance Indicators (KPIs) – Tracking relevant KPIs closely and reporting them regularly helps take away the guesswork from forecasting. Executives and decision-makers become more familiar with them – what drives them, when to worry if they shift – and as a result can become more adept at debating related assumptions for forward planning. By concentrating on evidence-based performance results for individual departments, Finance can help the business generate and set realistic longer-term goals and plans.

Debate and ‘book end’ key assumptions at Board – Executives should be encouraging regular Board level discussion and debate about key assumptions, stress testing scenarios, and exploring what unexpected events could significantly impact the business. For example, it’s unlikely many businesses seriously contemplated a total lockdown due to a global pandemic or factored its likely impact into their forecasting – but perhaps that sort of event might feature in more of today’s long-range contingency plans. Meaningful discussions, looking back and exploring the future, will not only help Finance to agree on best and worst case input values for forecasting assumptions (‘book-ending’ them) but should also help build organisational capability to withstand certain events and exploit potential opportunities.

Make (considered) use of AI – Artificial Intelligence can enable more extensive, faster, and sophisticated data analysis using multiple high-volume data sets – producing better results than more time-consuming and traditional data-gathering and processing exercises. By combining traditional approaches with machine learning and computer-based algorithms, AI seeks patterns in data. Its versatility allows for a wider range of information points to be gathered in real-time to predict future trends and consumer behaviours with increased clarity and data integrity. With good quality data, the scope for scenario generation and sensitivity analysis is much greater and this is a tool featuring more and more in today’s Finance functions. Consideration must be given to risks associated with its use – not least data security, sustainability impact (data centres have a high carbon footprint), and cultural factors – but it is undoubtedly a resource worth putting to work.

For now, it will be interesting to see how well AI has fared in the near term – though it would be even more impressive if it could make as good a set of predictions four years out!

Contrasting Approaches to Driving Sustainability