Relevance of Outliers in Pharma Sales Forecasting

Relevance of Outliers in Pharma Sales Forecasting

Just like any other predictions or forecasting – on a broader sense, pharma sales forecasts are also a tough and tedious task. Availability of many sophisticated state of the art digital software tools and techniques, notwithstanding.

In an article, published in the July-August 2007 issue of Harvard Business Review (HBR), Paul Saffo – a forecaster based in Silicon Valley, California – expressed this point succinctly with a nice example. He said: “Prediction is possible only in a world in which events are preordained and no amount of action in the present can influence future outcomes. That world is the stuff of myth and superstition. The one we inhabit is quite different—little is certain, nothing is preordained, and what we do in the present affects how events unfold, often in significant, unexpected ways.”

At this point, I would respectfully prefer to slightly alter the last sentence of the quote as, “….and what we and (others) do in the present affects how events unfold, often in significant, unexpected ways.”  This is important to me, as we may have control over what ‘we do’, but may not have much control over what ‘others do’ in the present, which may also greatly affect how events unfold, often in significant, unexpected ways.

However, the author distinctly differentiates predictions from forecasts by clarifying that prediction is concerned with future certainty, whereas forecasting looks at how hidden currents in the present, signal possible changes in direction for companies. Thus, unlike a prediction, a forecast must have a logic to it and the forecasters must be able to articulate and defend that logic.

My own hands-on experience in the domestic, as well as the pharma industry of the western world tells me that the actual sales and profit may seldom be a replica of the respective forecasts for the same. However. a reasonably good forecast is the one that is much closer to reality.

That said, it is important to note in the same context, what the above HBR paper has said, in this regard. The author underscores whether a specific forecast actually turns out to be accurate is only part of the picture. Citing a nice simile it says, even a broken clock is right twice a day. Thus, the forecaster’s one of the key tasks is to map uncertainty where our actions in the present influence the future. Uncertainty is an opportunity, he articulates.

In this article, I shall try to explore the possible reasons why, despite the availability of so many sophisticated digital software tools and techniques, the reality in most cases is much different. In a significant number of occasions, the actual sales is much less than the sales forecasts.

The criticality of forecast accuracy:

As we know, sales forecasts today are generally data pooling or consensus forecasts for better buying-in by the implementer, as there exists a critical need, just not to deliver closer to the forecasts, but to exceed the same, especially for the new products.

One will get the flavor of criticality of sales forecast accuracy from the McKinseyresearch study titled, “The Secret of Successful Drug Launches”, published in March 2014. It found that two-thirds of the sample group of drug-launches failed meeting pre-launch sales forecasts in their very first year on the market. The sample for this study comprised 210 new drugs launched between 2003 and 2009, for which McKinsey gathered necessary consensus-forecasts data for launch from EvaluatePharma. Three important findings of this EvaluatePharma – McKinsey analysis may be summed up, as follows:

1. Actual sales during the first year of launch as % of sales forecast one year before launch:

• % of launches below forecasts: 66
• % of launches on or near forecasts: 8
• % of launches exceeded the forecast: 26

2. Of launches that exceeded the forecasts in the year 1:

• 65% continued to do so in the year 2
• 53% of those exceeded forecasts in the year 3

3. Of launches that lagged forecasts in the year 1:

• 78%continued to do so in the year 2
• 70% of those lagged forecasts in the year 3

In an eloquent way, this study highlights the benefits of sales forecast accuracy for a sustainable performance excellence, especially with new products.

Wide room for improvement in forecasts:

Although, my focus in this article will be on sales revenue forecasts, there is a wide room for improvement in other related forecasts, as well.

Another interesting article titled, “Outsmart Your Own Biases”, appeared in the May 2015 issue of the Harvard Business Review revealed, when researchers asked hundreds of chief financial officers from a variety of industries to forecast yearly returns for the S&P 500 over a nine-year horizon, their 80% ranges were right only one-third of the time. The authors considered it as a terribly low rate of accuracy for a group of executives with presumably vast knowledge of the economy of the United States.

The study further indicated that projections are even further off the mark when people assess their own plans, partly because their desire to succeed skews their interpretation of the data.

Such a scenario prompts the need of yet greater application of a mix of creative and analytical minds to ferret out the reasons behind general inaccuracy in forecasting, which incidentally does not mean setting out an easy target, and then exceeding it. Right sales forecasting with high accuracy, is expected to make use of every potential future opportunity in the best possible way to achieve continuous excellence in performance.

Analyzing outliers in consensus forecasting:

A recent paper deliberated on this area backed by some relevant case studies to capture the relevance of outliers in consensus-forecasting for the pharma companies. Continue reading….