Agile forecasting

Agile forecasting

At the onset of the new year, companies are either refining their calendar year sales forecasts or about to start fiscal year sales forecasts. Many engage in three-year forecasting, some even extending to five years. Either way, these organizations spend tons of time and resources to create “accurate” forecasts. With the looming global recession, continued Russia-Ukraine war, disruptions in supply chains and unstable inflation, business forecasting has never been more challenging.


In my experience with global companies, there is always the tendency for management to “force” an optimistic business forecast for the current calendar or fiscal year to the tune of 10-, 20- or even 30-percent annual growth regardless of the macroeconomic forces. Senior management will then ask what resources operations will need to achieve the “high” forecast. This often results in overinvestment and underachievement, which then leads management to spend time and effort to cut down on resources such as the workforce and readjust the forecast instead of focusing on growth.

This practice of overpromising and underdelivering when it comes to business forecasts drains the organization of not only resources but credibility among senior management and other stakeholders. That is why it is important that forecasts be data- and expert-based as well as adaptive to changes in the internal and external environment. We call this approach agile forecasting, borrowing from the widely used agile methodologies in software development that are now also being used in functional areas such as marketing and human resource management.


The first step is the planning stage where data is gathered. Data is both quantitative (e.g., historical sales, industry growth, etc.) and qualitative (e.g., expert opinions from customers, salespeople, etc.). The planning group in the organization should make this a collaborative process, considering the expert opinions and acting as arbitrator between senior management who want an optimistic forecast and the historical data and salespeople inputs that may be on lower levels.

This brings us to the next stage, which is designing the forecast. A common method in uncertain times is scenario planning, where forecasts are classified into pessimistic, realistic and optimistic with corresponding probabilities of achieving. These probabilities are again derived in a collaborative manner, utilizing quantitative and qualitative means.

Once the forecasts are agreed upon, then they are deployed in terms of targets such as sales, productivity or output. This is where the practice of agile is critical. With regular reviews, e.g., weekly and monthly, corrective measures can be planned and implemented immediately. These may come in the form of reducing resources or holding off investments. Iterating the forecast regularly alongside recalibration of resources makes this a powerful approach to ensure “close-to-reality” estimates. But it is easier said than done.


Many global organizations employ these approaches but what’s lacking is the management will and mandate to listen to what is happening in the field and make tough decisions to reduce the business forecast. The integrity of the data used, be it quantitative or qualitative, is critical in justifying any adjustment in the forecast and convincing top management of the changes.

The author is the founder and CEO of Hungry Workhorse, a digital and culture transformation consulting firm. He is a fellow at the US-based Institute for Digital Transformation. He teaches strategic management in the MBA Program of De La Salle University. The author may be emailed at