Delivering an effective forecasting process
pharmafile | October 7, 2011 | Feature | Sales and Marketing |
It’s surprising how many companies struggle to deliver an effective forcasting process despite the millions spent on supply chains.
Accurate sales forecasts lead to lower inventory levels, fewer ‘stock outs’ and improved customer satisfaction. So why are so many companies struggling to deliver an effective forecasting process?
If you’re in sales you just need to concentrate on making target. And if you’re in manufacturing, make more, better quality products, at less cost. It’s the planner’s responsibility to make sure that there’s enough product ready to ship. As daft as that may sound, many global enterprises seem to have adopted these very principals, each entity working in their own private silo.
Improving forecast accuracy is essential to lowering your company’s inventory levels and reducing write-offs. Without essential input from the sales team to report accurate future demand from your customer, you are running blind. Only they can ensure that the right product is in the right place at the right time – which will improve customer service levels and reduce the need for fire-fighting and costly expediting. It’s an absolute necessity to achieving growth, increasing profitability (and maintaining any degree of corporate sanity).
Here’s a 10-point plan that highlights some of the most effective ways to increase the value of a sales forecast by improving on forecast accuracy.
- Forecast as close to the customer as possible
- Forecast demand, NOT sales
- Identify and separate different demand streams
- Modify the sales history to remove exceptions
- Remove distortions caused by promotions and supply limits
- Forecast at higher levels and ‘top-down’ to the detail level
- Separately manage records with limited history
- Collaborate and share forecasts and inventory to drive S&OP
- Make forecasting a key business step with clear accountability
- Measure performance and publish it to continuously improve.
Forecast as close to the customer as possible
A forecasting process is usually attempted wherever inventory moves from one point to another.
Obvious forecasting points would be:
- Customer to Consumer
- Distribution Centre (DC) to Customer
- DC to DC
- Plant to DC
- Suppliers to Plant.
The further away a forecasting process is from the end user, the more distorted that demand signal will become. Each step in the supply chain distorts the true picture. Despite widespread acceptance of this ‘bull whip’ effect, operations management teams way down the chain at suppliers-to-plants and plants-to-DC levels, ask for forecasting tools to improve their process. But why would you try and forecast something that you can accurately calculate? A raw material or component forecast should be calculated from the net replenishment forecast for finished products, which can be calculated from a sales forecast to a customer; taken from a forecast of consumer demand and built directly from point-of-sale data. The key is to calculate the base and then add the collaborative market intelligence from sales, marketing, distribution and manufacturing.
However, the general principal is: “the closer you are to actual product consumption, the more representative the forecast will be of actual product use”.
An accurate forecast of demand will generate more accurate plans all the way down the supply chain.
Base your forecast on demand – not sales
If you run out of stock and make no sales for several periods as a consequence, the sales history data stream would suggest that your customer had no demand for those months. Or, if the plant struggled to produce the quantities of what your customer wanted, when they wanted it, but was able to make and deliver one large batch, your sales history would appear to spike unnaturally. In both cases, using historical customer order demand by request date, rather than delivery date, will generate more accurate statistical forecasts of future demand.
After you identify exceptional behaviour, adjust the history of that stream to eliminate the exception. Iron out the kinks in the barrel before you take a shot at predicting future sales.
Identify and seperate diferent demand streams
Different demand streams can behave in differing ways. Online and retail sales channels can have very different demand profiles and forecasting future demand should be managed separately. In the pharmaceuticals industry, where total demand history by product may appear erratic and unpredictable, you may find the cause to be different demand streams. Separate the lumpy and unpredictable tender business from the more regular, seasonal and statistically forecast-able pharmacy and distributor business.
Use forecasting algorithms to accurately calculate one stream and use collaborative tools to encourage the sales team to provide the best and latest ‘market intelligence’ for the other.
Understand and modify history to remove the effect of exceptions
Any good forecasting package will have the capability to highlight and filter out exceptional circumstances so that history can be ‘cleansed’. This allows forecasting professionals to remove those one-time variables that can skew a forecast.
Typical factors that distort history include:
- Promotions
- Stock outs, write-offs and ‘Fire sales’ to move inventory
- Unusual competitor activity
- Parallel imports
- New entrant buying market share by undercutting prices.
Identify and remove distortions of demand and supply parameters
- Batch size and max/min order quantities may be essential for logistics but these planning parameters should be excluded from any prediction of future customer demand.
- Preset distribution intervals that lack a direct connection to customer consumption will also diminish forecast accuracies, and their effect should be normalised.
- Discounts for increased order size may boost cash flow and one month’s P&L but they distort demand.
Never forecast what you can calculate
Forecast at aggregate levels and explode to lower levels
An effective forecasting tool will allow users to forecast at aggregate levels.
This technique will often produce forecasts of much higher accuracy as long as the groupings or hierarchies are meaningful and manageable. Sophisticated forecasting packages can allow forecasts at the total level to be automatically exploded to individual SKUs.
Identify and manage records with limited history
Identifying and managing records with short history is critical. Future sales of products with just one month’s actual sales are difficult to predict. Short history sales data should be reviewed each month.
As history accumulates the demand patterns will start to become clearer but products with fewer than 12 months historical demand need to be reviewed regularly. Using the demand curves of similar or related products to predict sales for new products is a good policy.
Publish forecast accuracy results
Communicate, share and use the data to drive your S&OP process
Having multiple forecasts with differing information throughout a supply chain creates waste, confusion and frustration. Forecasting true demand is the first step to achieving ‘one set of numbers’ and is essential for a business-wide sales and operations planning process. This goal is difficult to achieve using static spreadsheets to manage all of the data that drive activities from product development to delivery.
The multiple spreadsheet approach breeds silo thinking where everyone has their own, skewed version of the truth. One set of numbers, derived and agreed through collaboration on a web-based forecasting system that integrates with your existing data systems, will provide a focus for improved forecast accuracy and business efficiency.
Make forecasting a key business step with clear accountability
Sales forecasting must be a regular and highly visible process that is performed efficiently and rapidly. It should be seen as a key value-added business process to help deliver business control and execution of the commercial actions agreed in sales and operations meetings. Effective forecasting will only happen with a clear assignment of responsibility and accountability. Usually these rest with the sales and marketing team because this group:
- Can update the forecast with actual sales information immediately
- Knows where to modify history to deliver a better statistical forecast
- Knows where the advertising and trade promotional spend will go
- Knows and understands the short, medium and long-term brand performance
- Performs market segmentation and profitability analysis, not just historically but into the future as well
- Requires the resulting performance data for aggressive portfolio management.
“What gets measured gets done.”
Measure performance, publish it and continuously improve
Measures that should be part of any company’s balanced scorecard are sales forecast accuracy and forecast bias. It takes discipline and courage to publish the first and subsequent sets of forecast accuracy figures so that a company can chart continuous improvement.
Every month you should take a close look at the ten ‘biggest misses’. Understand what went wrong and use this to deliver a better set of forecasts next month. Don’t worry about what your forecast accuracy percentage should be.
The key to success is to start measuring. The number will change as you get better. Use the techniques above and continuously improve.
REFERENCES
Demand Solutions. Management Series White Papers (2010)
Solutions 4 Planning. Market Watch White Papers (2009-2011)
For more information please visit:
http://www.solutions4planning.com
