Category Archives: Forecasts

How Not to Excel at Forecasting

This post originally ran four years ago. But since a critical mistake is still being made, it’s time for a repost.

How Not to Excel at Forecasting?

Simply put, use Microsoft Excel. It’s appalling that a survey by ToolsGroup and the Global Market Development Centre (GDMC) found that even though two-thirds of companies in the consumer goods supply chain consider demand volatility and forecast accuracy a high businesses priority, half still rely on Excel spreadsheets for forecasting.

Relying on Excel for forecasting is like relying on:


  • a Longship to get you across the Atlantic

  • your first guess on Let’s Make a Deal to be the right one

  • a shareholder proxy getting on the ballot at a Fortune 500

  • Florida surviving a hurricane season without any major city suffering damage

  • the price of fuel going down and staying down for an upcoming series of spot buys

  • natural resource supply to be consistent and predictable year-over-year

  • a flip of a fair coin to come up heads seven times in a row

Now, it’s true that:


  • the Vikings did make it across the Atlantic in a Longship, but a single storm could sink it

  • the first door you pick, with one-in-three odds, could be the right one, but the odds are actually twice as good if you switch

  • an activist shareholder can sometimes get a proxy on the ballot if he or she has enough time and money, but as pointed out by John Gillespie and David Zweig in Money for Nothing (How the Failure of Corporate Boards is Ruining American Business and Costing Us Trillions), examples are few and far between

  • even though no storms made landfall in Florida in 2011, this is Not a common occurrence

  • gas prices did consistently drop in the USA between September 2008 and December 2008, but have been otherwise steadily rising for the last five years

  • in some years the rice, sugar, and corn crops are almost the same as in the previous year, but given the increase in hurricanes, tsunamis, droughts, and other natural disasters in recent years, this is not a common occurrence

  • yes, heads can come up seven times in a row when flipping a fair coin, but the chances of this happening are less than 1%

In other words, you can forecast with Microsoft Excel, but your chances of doing well, especially given that 90% of spreadsheets have non-trivial errors (and collectively cost enterprises billions, as Fidelity and Fannie Mae found out), are (vanishingly) small (as the complexity of the forecast increases). One has to remember that there’s no intelligence behind a spreadsheet and they are just a source of peril that can cost your organization millions without anyone noticing.

Consumer Sustentation 74: Demand Planning

Demand Planning is a damnation. Why? As per our original damnation post,

  • traditional demand planning models require historical data
  • traditional demand planning models require market predictability
  • traditional demand planning models require market foresight
  • traditional demand planning requires knowledge of the expected price point

And how often in today’s constantly changing consumer marketplace, with new product releases coming faster and faster (to the point where your phone, laptop, and music device is out-of-date by a whole new release within a year), do you have good historical data, market predictability, and foresight? And how often can you be confident in the price-point, as a skunk-works product release by a competitor between sourcing and sale can force a price reduction to prevent inventory sitting on the shelves indefinitely.

So what can you do? (Besides burying your head in the sand like an ostrich?)

1. Get as much market data as you can.

Collect as much data as you can on your competitors imports, sales, and revenue using publicly accessible import data, analyst data, and company annual reports. It won’t be accurate, but with enough data you can often identify better trends than you could on the most similar product in your own inventory (which might not be similar, or recent, enough to be sufficiently relevant).

2. Have third parties conduct surveys on your behalf.

Sometimes the best way to gauge a market forecast is to actually conduct customer surveys and have a third party use the data to estimate demand for you. If you have no clue, the best thing you can do is admit it and get an expert to help you come up with a realistic demand forecast range.

3. Don’t focus a number, focus on a range and a potential rate of ramp-up or ramp-down.

If you know the demand is expected to be in the 100K to 200K units a month range, and the demand could double overnight, then you know that you need to contract for the low-end, but with a supplier that could ramp up to double production in a matter of weeks if necessary. And you have to negotiate a contract that allows orders to escalate, with pre-defined increases if the supplier is forced to work overtime (so you don’t get any billing surprises or animosity down the road).

4. Keep on top of sales data in real-time.

Be sure to get at least weekly PoS updates, and re-run the projections on a regular basis to detect an upswing or downswing early, so that you don’t get caught with your pants down, or, even worse, your pants off.

If you follow these tips, then you can get a reasonable grip on demand planning while your competitors flounder with the flounders.

Consumer Damnation 74: Demand Planning

Each group of customers are a damnation upon themselves, and they will get the attention they deserve, but demand planning to meet customer demand is its own damnation. Why is this?

Traditional demand planning models require historical data.

To be precise, they require a fair amount of historical sales or usage data in order to be accurate. And sometimes a lot of sales data. But with new product introductions coming fast and furious every day, there are so many categories without a decent amount, if any, historical sales data that it’s hard to make good predictions. Now, one can always use the most similar product, or the product the new product is expected to replace, but this weakens the model and the confidence in the result.

Traditional demand planning models require market predictability.

To be more precise, they expect that the market will not substantially change. That the needs will stay the same. The utilization or replacement curves will stay about the same. That a competitor won’t substantially increase or decrease their market share overnight. That a revolutionary new product won’t be released that causes a huge market shift.

Traditional demand planning models require market foresight.

In addition to requiring historical data and market predictability, traditional demand planning requires market foresight. Knowledge of potential competitor product introductions that could change the market demand. Knowledge of innovations that will begin demand shifts. Knowledge of general market conditions that could delay replacements or result in reduced demand due to cash availability.

Demand Planning requires knowledge of the expected price point.

Most products are services, especially in the end consumer market, are very price dependent. People will pay more if they perceive more value, which could be better quality, more functionality, or owning an iconic brand, but if they don’t perceive more value in your product which is priced higher than a competitor’s product, don’t think for a minute, even if they bought from you last time, they won’t shift. And price prediction is difficult if it is dependent on production cost, which can be variable if transportation can involve unpredictable fuel surcharges, raw material prices can skyrocket due to insufficient supply as a result of a disaster, and labour prices are dependent on contingent labour to meet demands at peak periods.

In other words, sometimes demand prediction models fall flat, and demand projections come from a place that can only be seen by a proctologist with a flashlight, so how do you effectively plan for those as a Procurement Professional? You don’t. It’s damnation.

Finally, a Prediction SI Can Get Behind!

By now, everyone should know how SI, and the LOLCats who live under the desks, feel about futurists and their predictions. (You need only scroll back to December 31’s post if you have forgotten.)

So, needless to say, as per prior years, SI is not going to be jumping on the prediction bandwagon (and risk getting trampled by fellow bloggers on the way) as the new year rolls in.

That being said, it has to give a shout out to one prediction from a fellow blogger who may just have it right. Specifically, Peter Smith of Spend Matters UK who, pressed for a prediction, made the amazingly accurate prediction that we predict that all predictions will be wrong.

SI could not have said it better if asked.

LOLCat approves!

Procurement Trend #26: Increased Accuracy in Demand Planning

Twenty-three lacklustre, backwater, trends from yester-year still remain, so let’s get back to it. The sooner we get through these, the sooner we get back to modern times.

So why do so many historians keep pegging increased accuracy as a future trend, and helping poor LOLCat regress to past lives? There are a number of reasons, but among the top three today are:

    • Hyper-competitive markets make demand planning difficult
      because a one week’s difference in release date due to an unexpected delay can result in a competitor beating you to market and siphoning off a significant portion of your expected market share for the product
  • lack of long-term data in short lifecycle product categories makes trending hard
    which makes it even harder to predict not only when a product instance is going to reach end of useful (sales) life but when the next iteration is going to bomb because the product has reached end of life and needs to be retired
  • most tools are still using outdated inventory models
    because the initial versions were created twenty, thirty, and even forty years ago and it’s just not possible to force fit new, complex, innovative inventory costing and projection models into them

So what do you do?

Hyper-Competitive Markets

As per above, Procurement not only needs to identify suppliers who can add value at little or no incremental cost but needs to identify suppliers who can help it get an edge in these markets. It needs to move to JIT (Just in Time) production and distribution to the extent possible, track product and consumer trends carefully, and adapt as needed.

Lack of Long-Term Data in Short Lifecycle Product Categories

It needs to collect as much market data as it can from analyst and trade bureaus to identify global trends, and analyze all of the data it has on past and current products to predict life-cycle trends that are in-line with current market conditions.

Outdated Inventory & Forecasting Models

It needs to update its inventory management and demand planning tools ASAP to not only plan with more data, more resolution, and more options, but support forecasting under different conditions.