The factors that influence forecast accuracy
In 1982, Mark Moon, John Mentzer, Carlo Smith, and Michael Garver mailed 157 companies surveys to understand how they created their forecasts. By 1998, they had studied the forecasting practices of hundreds of companies, going far beyond survey questions and into benchmarking studies, in-depth analyses, and forecast audits.
The team published their findings in a surprisingly accessible 8-page paper titled “Seven Keys to Better Forecasting.” Today, we will attempt to compress their findings even further and contribute our own consultancy experience by adding an eighth key to the forecasting key chain.
Forecast success factors
Moon and Mentzer’s team concluded that sales forecasting is a vital management process that companies must prioritize. Forecasting should focus on predicting actual customer demand using both quantitative and qualitative techniques. Companies should establish processes and systems for both producing and assessing forecasts. Setting up an independent forecasting department with the power to bring different organizational areas together facilitates the creation of a consensus forecast, as does training everyone involved in the forecasting process. As these practices are implemented and forecasts improve, companies can expect to enjoy reductions in cost and increases in customer and employee satisfaction.
1. Understand What Forecasting Is, and What It Is Not
Sales forecasting is a management process performed by humans, not a problem solved by software. A sales forecast is an estimate of what future sales might be, given certain environmental conditions. It is distinct from a sales plan, which is a commitment, as well as from a sales goal, which is a target.
2. Forecast Demand, Plan Supply
Forecast customer demand, not your company’s ability to supply goods or services. Using historical shipments to generate forecasts will cause you to repeat your mistakes. Identify where capacity doesn’t meet demand forecasts and start the process of expanding. Forecasts should be based on unconstrained sales: the sales that you would achieve if you had no capacity constraints or stock-outs.
3. Communicate, Cooperate, Collaborate
Move beyond just communicating or cooperating, instead aiming to foster equal collaboration among all functional areas. Consider creating an independent department with the power to bring different organizational areas together to create a consensus forecast.
4. Eliminate Islands of Analysis
Establish a process for producing a forecast all departments trust. Include software that communicates seamlessly with other information systems, a suite of statistical techniques and graphical programs, an ability to capture and report performance metrics over time, and a centrally-maintained, accessible “data warehouse” with historical sales data. Then teach employees and developers how the process works, their role in it, and the importance of forecasting.
Our perspective: We’ve seen that a lack of trust in forecasts is a chronic problem for many companies. We can usually lower a company’s safety stock levels by about 30% by challenging unproductive adjustments made to forecasts.
5. Use Tools Wisely
Combine quantitative and qualitative techniques. Implement a process that uses time series to forecast trend and seasonality, regression analysis to forecast demand relationships with external factors, and qualitative input to adjust these forecasts. Train key personnel to assess and use forecasting techniques and make adjustments to forecasts. Then refine this process by finding which techniques best improve accuracy for your company.
Our perspective: We provide advanced forecasting applications and processes, along with the training clients need to use them. Our consultants regularly prompt the appropriate teams for feedback.
6. Make It Important
Train everyone involved in the forecasting process. Educate forecast developers on how their forecasts are used, and other employees on the needs of forecasters. Incorporate forecasting performance measures into job performance evaluation criteria for senior managers and front-line employees. Provide feedback and rewards for positive results.
Our perspective: Our experience shows that working with forecasting consultants on an ongoing basis motivates clients to prioritize forecasting. Training, evaluations, and feedback then motivate the clients’ employees.
7. Measure, Measure, Measure
Implement systems for measuring performance, tools for providing feedback, and standards and targets for what constitutes forecasting excellence. Ensure that forecast creators review their history and assess their performance. Performance metrics should reflect the measures used by different departments to assess accuracy. Track accuracy at each point at which forecasts may be adjusted.
Our perspective: We practice what we preach by documenting the financial benefits of our tools and consultancy services on a monthly basis. Clients are free to terminate our relationship should we ever fail to live up to their standards.
And dont forget to…
8. Manage Planning Behavior
To improve forecast accuracy, start with planning behavior . Question assumptions, traditions, and incentives. Challenge planning decisions. Identify and address counterproductive habits. Follow up frequently to maintain efficient.
Great forecasting is what you get when you bring together advanced software, well-crafted processes, and employees from all levels and departments collaborating in a spirit of democracy. These employees must be supported through training, a consensus-based discussion environment, and regular evaluation in order to excel.
Insights like this reach far beyond forecasting. Nearly everything we enjoy today, from Internet access to commercial flights, comes from teams of people using software and processes to do things one human could never do alone. We learn in kindergarten that we need to be shown how to do new things, that we should work together and fairly in teams, and that being evaluated helps us do better. These lessons are taught early because they’re both vital to our survival as a species and difficult to follow consistently.
Having served more than 50 companies across industries and regions, we at Perito Consulting have seen teams become stronger overall as they improve their forecasting capabilities. We are proud to facilitate this growth. The quantitative benefits of working with us can be measured through figures like a 30% increase in forecast accuracy. The qualitative benefits, like an increased spirit of teamwork, are just as powerful.