Dynamic business and financial forecasting models

While it's impossible to accurately predict the future, our job as a leader requires us to make forecasts. Every day, leaders across your organising are asking questions like:

• What will happen in the economy?
• What will our competitors do?
• What will our customers want?
• What will happen if we introduce a new product line?
• What will happen if we change our pricing?

The answers to these questions help us navigate a way forward. Sometimes our forecasts will be right, and the choices that we make will create value. But sometimes our forecasts and choices will be wrong and we will destroy value. To stack the odds in their favour, leaders rightly spend significant time and resources to improve the accuracy of their forecasts.

What most businesses do

For most businesses and leaders, forecasts are constructed using a combination of intuition, workshops, financial models, market research and historical data analysis. While all of these methods have their merits, they fall short because they fail to capture the interconnectedness and sophistication of not just each part of a business, but the business and it's competitors, and the industry and the broader economy.

‍

For example, consider the following limitations of:

• Intuition - can any one leader truly understand or comprehend the drivers of success for the entire organisation? The CEO doesn't know what's happening operationally, and leaders closer to the frontline don't fully understand the strategy
• Workshops - can we cover the blind spots of an individual by working as a group? Maybe, but without a well-defined structure, workshops usually fail to achieve consensus. While interesting for all involved, they are often of little practical value
• Financial models - often viewed as the definitive 'quantifiable' method of forecasting, a financial model can tell you - down to the dollar - what your future performance will be. The critical shortfall of financial models is that they don't capture the true business drivers and dynamics, and as a result the questions they ask are simply too high-level. For example, an assumed growth rate on revenue isn't that helpful when you consider the how many products there are, price levels, demand and price sensitivities, etc., which are not considered by a financial model
• Market research - what our customers want and what they are doing can certainly help us ask the right questions. But market research is of limited benefit in deriving answers because they are by definition externally focused, and are completely separate to the mechanics of the business and its strategy
• Historical analysis of data - like market research, this can also help us ask the right questions. But assuming that the past is a reliable predictor of the future is a dangerous assumption 

Taken individually, each of these methods are simply insufficient to give us any level of confidence in our ability to forecast. In reality though, most businesses will rely on a combination of methods and hope that they've done enough to make accurate predictions. 

‍

But what if there was another tool at our disposal that could make our forecasts even better? What if this tool actually built on and consolidated all of the methods that businesses are already using? What if there was something that could give us an unfair advantage?

What smart businesses do

A business model is just that tool. A business model, as distinct from a financial model, is a dynamic and sophisticated model that rebuilds the mechanics of your business and allows you to perform detailed what-if forecasting. A good business model will have enough sophistication to represent the key interdependencies, dynamics, and decisions in your business without creating unnecessary complexity (which is both costly to build, and can cloud your ability to draw inferences from the results).

‍

The key characteristics of a business model are:

• Each of the business's divisions and their interdependencies are includedUsers are given the opportunity to modify a set of strategic assumptions
• Each of these strategic assumptions are used as inputs that feed the logical mechanics of the model
• The mechanics are fed through a combination of assumptions and known data points. Known data points are likely to replace the need for assumptions as the model mechanics become granular (broadly speaking, strategic = assumptions, operational = known data points)
• The integration of a financial model, which effectively becomes the output of all of the decisions, assumptions and model interdependencies
• An intuitive interface that provides executive-level insights and an easy to configure dashboard

A business model - known data points, business mechanics, and assumptions - can then be informed by all of the 'traditional' forecasting methods. The model can be built from market research and the historical analysis of data, layered on top of a financial model, and then the assumptions can be rapidly tested and refined through executive intuition and workshops.

Our approach

We believe that there is no one-size-fits all approach to building a great business model. The value of the model lies in its ability to help your leaders make the best decisions. Each industry is different, each business is different, and at any point in time each business will need different insights. Our approach is two lay the foundations during the first phase, and then world in modules or sprints as needed, which incrementally extend the functionality of the model.

‍

During phase 1:

• We take the time to understand how your business makes money, it's strategy, what success looks like, and the business and project's key constraints
• A basic three-way integrated financial model is built, mirroring the real-world financial statements
• The core building blocks of the business model are created to create conceptual integrity (linking cost of goods sold to revenue, building asset profiles of staff and key property, plant and equipment, etc.)

Phases 2+:

• The model is extended on an as-needed basis. Triggers could be a new product launch, the desire the model an economic shock, or the need to peform a sensitivity analysis on a business case under consideration
• The extensions built during each phase is integrated with the existing model. Over time, this provides an increasingly robust and realistic model