Marketers of all stripes are often tasked with forecasting–sales for next quarter or year, inventory levels to meet demand, or marketing budget to meet corporate goals. However, the process of forecasting is often rife with bias, data quality issues, mathematical error, and/or poor planning assumptions. While no forecasting technique is perfect, predictions can be drastically improved through a simple technique: pulling your anchor.


Let's face it, the ability to peer around the corner and forecast the future (tomorrow, much less next month) is one of the biggest challenges for global business executives. If you have made forecasting mistakes in the past, take heart, you're not alone. You can however, learn to forecast a bit better by avoiding a common mistake–anchoring.

The concept of anchoring in decision making was made famous by psychologists Daniel Kahneman and Amos Tversky. In a well known experiment, they asked a group of people to estimate the percentage of African countries in the United Nations.

Before each person guessed, they were asked to spin a wheel with the numbers 0 to 100 clearly labeled. According to a Money magazine article, "When the wheel landed on a low number, people tended to guess that African nations made up a small percentage of UN members; when it landed on a high number, they guessed that Africa accounted for much more of the UN's membership."

A recent Wall Street Journal article also highlights the problems of anchoring. The article mentions that in December 2008, Barron's asked a dozen experts to forecast the level of the Standard and Poor's 500 stock index at the end of 2009.
Despite a tumultuous 2008, where the stock market would regularly jump 500 points in a single day, not one expert predicted a down year for 2009. And while 2009 still has yet to play out, each of the twelve experts predicted a 5-38% increase with a median of 13%!

The author of the WSJ article, Jason Zweig, believes the stock picking experts are guilty of anchoring. He says we tend to over-extrapolate in forecasting when we base our decisions mostly on what happened in the past. Zweig writes, "(With) the fat five years from 2003-2007, when stocks shot up by an annual average of 12.8%, who expected 2008 to be a bloodbath?"

Who indeed? With the exception of Nouriel Roubini and a few others, most market prognosticators missed the mark.

In fact, accurate forecasting can be hard–or easy. Let me explain.

Years ago, as a divisional manager for a regional telecommunications firm, I was responsible for presenting my annual division revenue forecasts to the company president. These forecasts were very important, as staffing decisions and budgets would be initially based on the "acceptance" of the forecasts.

When visiting the office of another divisional manager, I was shocked to find out he was finished with his forecasting processes–in record time. He finished quickly by taking last year's revenue numbers and adding 10%! Adding insult to injury, his last year plus ten percent forecasts were accepted by the company president!

Alas, the above story would be much richer if my colleague's exercise in anchoring ultimately cost him dearly. He was lucky–in this instance his predictions came pretty close to reality. His gut decisioning (or perhaps laziness) came through for him.

However as levels of volatility and systemic failures increase in marketplaces and economies, and wild swings become the norm–anchoring and basing forecasts on what happened last year can ultimately lead to disaster. Need proof? Just query the terms, "investment bank" or "hedge fund" in the search field of the online Wall Street Journal.

To avoid anchoring in decision making, carefully consider assumptions and take no probability off the table. Think about your "reference point" for making a decision and then consider reasons why this particular point is your anchor. Does it take into account the possibility of an extreme outcome?

If you think it couldn't happen–pay special attention to that scenario. It's pretty wild out there–anything is possible!
Questions for DailyFix readers:

* In a tough economic environment, the penalty for poor forecasting increases. What tools and methods are you using to anticipate events and predict the future?
* Author Peter Bernstein says, "The successful businessperson is a forecaster first–all other activities follow." Do you agree with this statement?
* What "experts" via news, commentary, consulting etc, are helping to guide your decisions on the future? Are these experts getting more or less reliable?
* What is the one event–if it happens in 2009–that will shake up the game board and change all the rules? Is this event on your radar screen?