Forecasting: Science or Art?

Demand forecasting is the holy grail of inventory control.  The value of demand forecasting is almost axiomatic – it doesn’t really require a complex modeling exercise to prove it.  However, corporate America is filled with stories about the forecasting pitfalls and problems.  While there have been great discussions about various algorithms used to forecast demand, let’s take a fresh look at forecasting.  Is forecasting science (defined algorithms) or art (human judgment and intuition)?

Forecasting algorithms are actually very well known mathematically, and it seems that graduate students in Operations Research (an area of applied mathematics) are coming up with new ones on a regular basis.  While a time-series type approach may work for established products, other algorithms would be needed where there is little historical information or where the forecasted event is intermittent. 

While algorithm performance varies depending on the situation  (for example, exponential smoothing and double exponential smoothing work differently), any of them will outperform simply allowing the ordering process to run on autopilot.  The mathematics behind these algorithms are quite elegant and provably correct.  However, given that is the case, why did Nike has such a spectacular demand forecasting failure in 2004?

Science Meets Art

In the case of Nike, court documents demonstrated clearly that Nike relied too much on software to do the forecasting, rather than on people who were clearly part of the supply chain process.  As one practitioner put it, you can always assume that the initial forecast will be wrong, especially if the algorithm is being fed bad or out-of-date data.  So, let’s look at a couple of ways that forecast accuracy can improve.

The law of large numbers basically says that the more you have of something, the more accurate are your forecasts.  Life insurance companies are so profitable because they have large pools of people, segmented by demographic and health factors, and can with a high degree of reliability predict how many will die in a given period.  Most insurance companies exclude acts of war because that is something they just can’t model in their forecasts.

So, the first step is to aggregate the demand into batches large enough to increase the forecast accuracy.  For example, let’s say there is a new product launch nationwide, and enough product needs to be ordered to meet demand.  What are some steps marketing can take?

There are factors which include seasonality, point in the product lifecycle, weather, and any number of other data.  It is difficult to provide an exhaustive list, but the point is that forecast accuracy improves as factors are introduced which more closely models reality.

Art Meets Organization

Aggregate demand is made up of segmented demand, and it’s a judgment call at to which customer segments make sense for a particular forecast.  Marketing will have already done research on the expected market for a new product, and the customer segmentation they expect to materialize in the marketplace. 

Sales will have a big say in whether the forecast is reasonable, since they will have to sign up for the number.  If marketing’s forecast says there is demand for 1,000 shirts, sales needs to commit to finding the demand and selling into it.  They will look at some of the same data as marketing, but will also work their sales channels to see if it can move that much merchandise. 

Once sales is done with it, finance and operations will need to weigh in to make sure they can finance the inventory and the infrastructure exists to move that many shirts.  A trade off is possible between the addressable market, the available market, and corporate resources.

As the foregoing implies, there is a process for getting to a forecast that ultimately turns into a sales order.  The technology and original forecast was balanced against the organization and what it could accomplish.  It is a balancing act, as has been discussed elsewhere. 

Placing the Factory Order

All of this activity leads to the factory order.  Once the order is placed, the challenge shifts from inventory size to inventory management.  That will be a discussion for another time. 

So, is forecasting an art or a science?  Every forecasting situation is different because each product or service is different.  Having said all that, forecasting is really art buttressed by science.  Put in place processes and so forth to help the people use the science to generate good forecasts.  The payoff is worth the effort.

Technorati Tags: demand forecasting,forecasting art,inventory control,demand planning,forecasting problems

CRM: Building the Business Case

Customer Relationship Management (CRM) seems to have a somewhat checkered history of delivering the expected value.  Sometimes, the reasons for failure are the same as those for other IT projects – project management, budget issues, etc…  The issue I’d like to address here is expectations as it relates to the results of a CRM initiative. 

Early in the project, the sales or internal support team will begin to formulate a selling proposition (internal teams need to be good sales people too), which will include a business case for the project.  Often, either the business case is weak (“You will experience 5-10% improvement in operations”), or it is overly-optimistic (“You will have a 15% increase in passenger traffic on your airline”).  Both lack credibility.  Being credible requires a clear line between the expected benefits and project-linked improvements.

So, here are some suggestions that will help create stronger business cases that are compelling and set the foundation for success.

• Numbers.  The most believable aspects of a business case are always the numbers.  “Improving” and “improving by 7%” are very different.  Anyone can put a number on a PowerPoint slide, but fewer can back those numbers up with a detailed analysis. 


• Story.  The story associated with the numbers are probably more important than the numbers themselves.  Mark Twain said, "Figures don't lie, but liars figure."  That truth is not lost on clients, so the story must be credible.  What makes a story credible?  Read on.


• Integrity.  At some point, you’ve gotten good data to create a credible set of numbers, and you’ve put together a story about those numbers.  It now comes down to you – are you credible?  You must believe your own story, and you must be fluent with its presentation.  That flows from your own integrity.

These steps are not linear but rather occur simultaneously.  Your story will guide your search for relevant statistics, and those same statistic will guide the development of your story.  The project itself will put you in a box (e.g. project constraints), which in turn influences the set of numbers required.

For example, in a given customer engagement, I needed to put together a cost-savings business case based on an application change.  The first choice was deciding how to explain the business case.  Do you try doing a comparative analysis between the incumbent and new software solutions?  The problem with that approach is fans of either package start to compete with each other and it blunts focus.  Thus, the first rule is: keep the story simple.  I chose to focus on the relative development and operational costs.

Next, gather as much relevant data as possible.  Especially early in the cycle, getting reliable data is often difficult.  Clean data is great but often hard to produce.  Get as much clean data as possible. 

How many points of value should you create in your narrative?  The answer brings us to the second rule: compelling is better than exhaustive.  Demonstrating 80% of the value create is not 100%, but if telling the story with 80% only requires five value points, versus 25 for 100%, I’ll take the five every time.  The story becomes compelling when it becomes easily comprehensible. 

Finally, sequence the story as a series of reveals and get to your point quickly.  While you still have to step through the story, don’t take a long time to get there.  If you’re using PowerPoint, a few slides should be sufficient. 

There are a lot of “it depends” factors in this analysis: size of the investment required, audience, criticality of the benefits, etc…  Getting the numbers, crafting and presenting the story, and then acting with integrity are key to being believable and setting a realistic expectation early in the project.  If the project starts well, it will end well.  By the way, in case you were wondering, the story I told earlier resulted in the proposed solution displacing a good incumbent provider.  That outcome was achieved because the client believed the business case.