A Cautionary Tale about Forecast Inaccuracy

An excerpt from my new book due to be published at the end of the summer

Prediction is very difficult, especially if it's about the future! —Niels Bohr, Danish physicist

Night of the Long Knives

I came into work one morning to learn that most of the senior management team had been fired. In the following few months, everything was different at the company where I had worked for five years. The most tangible change was that the company closed the plush corporate office where I had worked. They gave me a choice: relocate to the manufacturing site or adios. Given that I had a wife and three kids settled in a lovely small town on the water, with many good friends, there was no desirable choice. In the end, we reluctantly relocated from the San Francisco area to be close to the manufacturing site in Southern California.

One of the key reasons we decided to move was that the company, which had previously been hemorrhaging cash (hence the firings) had just acquired a major new customer. This account promised to be the savior of the company with 1,000 stores across the country. In fact, the sales team projected that this new customer would increase our total revenue by 30 percent. As you can imagine, there was a new excitement and positivity that had been missing for some time.

On my last day at the old office, the CEO shouted after me “Don’t screw it up!” as I packed my belongings. Since I headed the US Supply Chain organization, it was my responsibility to ensure that the new customer got everything it needed, on time and in one piece. No pressure, right?

Everything worked well for several months. Then one day, the sales team sent me an updated forecast as they did at the start of every month. Typically, I would start by quickly scanning the new numbers to see if any of them stood out. There would be some minor changes, but nothing to get excited about. Not this time! I will not repeat the words that instantly left my mouth.

3X Forecast Increase

Bill was the sales manager for the new account, and he had suddenly tripled his unit forecast starting in two months for one of the major product lines. We manufactured and marketed ceramic products for the bathroom (i.e., toilets, sinks, etc.). In other words, production output would also need to triple very quickly to meet this enormous increase. Was he smoking something? I gave him a call, and he confirmed that the numbers were accurate. It was a promotion he had been working on for some time, but he did not want to say anything until he had sealed the deal. Thanks, Bill! I knew this would be the start of a long few months, as we worked tirelessly to quickly achieve a 300 percent increase in production.

After I picked myself up off the floor, I dusted myself off and tried to think. We had outsourced this product to a contract manufacturer in Colombia. Their standard lead time was three months, and this did not account for the four-week transit time to the US. I knew my bosses would not receive the news well if I told them that tripling production in the time required would not be possible. Even though it was incredibly bad communication from the sales team, management would still expect the supply chain group to deliver. Put it this way: the company had not yet onboarded the notion of Don’t shoot the messenger. It was especially true in this case, due to our financial challenges at the time.

Knowing that “no” was not an option, the only solution to this massive challenge would be to import products by air instead of ocean. This would trim the transit time from four weeks to two days. As you can imagine, given the bulky nature of the product line, the elephant in the room was the cost of this alternative shipment mode. It would be several times more expensive, but we had no choice. So, that is what we did. We contracted with an airline and chartered planes to import the products from Colombia to the US. If this sounds like an episode of Narcos, the thought did cross our minds that this could look like suspicious activity.

Mothers’ Day Flowers

For several weeks, it seemed that we had dodged the bullet, and we were able to deliver on time. That was, until I received THE email.

Dear Mr. Clarke,

Thank you so much for your business. We regret to inform you that there will not be any aircraft available for the next few weeks. At this time of year, we allocate all our aircraft for importing roses into the country for Mother’s Day.

Regards,

Joe

I can still remember the gut-wrenching feeling. I knew that the precious new customer could well be out of stock for its huge national promotion within the next couple of weeks. This would be a prime example of screwing it up. When I read that the reason for the aircraft shortage was due to doting children and husbands buying flowers for Mother’s Day, I didn’t know whether to laugh or cry.

My first instinct was that I should update my resume. Then my next thought was, how could I explain this to my family? “Hey guys, you know how I just uprooted your lives for my job? Oh, and not to mention that you left all your friends behind. Well, funnily enough, they just fired me, so all the upheaval was for nothing. Life is funny sometimes, isn’t it guys?” My wife would be more understanding if I explained that it was for a worthy cause. But then again . . . .

Fortunately, however, we were able to persuade the contract manufacturer to temporarily raise their capacity, which somehow just kept our heads above water. My worst fears did not happen, and I retained my job. In fact, we recovered so well that we received a commendation from the customer for our outstanding delivery performance. Thankfully, they could not see behind the curtain.

Forecasts Are Important

The reason I tell you this story is not just because it is quite amusing (at least, in retrospect), but also to highlight that forecasts are important[CM1] [SC2] . In fact, in his 2017 paper “Win the Business Case for Investment to Improve Forecast Accuracy,” Gartner practice vice president Steve Steutermann found that for every 1 percent of forecast accuracy improvement, companies on average realized gains including 7 percent reduction in finished goods inventory, 2 percent reduction in transportation costs, and a 9 percent reduction in inventory obsolescence.[1]

Just think about how important forecasts are in everyday life:

·         We get wet when it rains unexpectedly.

·         We go broke if we spend more than we budgeted.

·         We lose savings if the stock that we fancied turns out to be a dud.

·         We become sick if approved medications do not work as predicted.

·         We wait at traffic lights if traffic flow in the city center is not as projected.

The bottom line is that terrible things happen when the crystal ball breaks. Even though they will never be perfect, there are many ways that we can make forecasts better align with reality.

3. Steven Steutermann, “Win the Business Case for Investment to Improve Forecast Accuracy,” Gartner, 2017.