‘Breaking Failure’ - Using Analysis and Strategy
Originally published: 2016 (PDMA Visions Magazine • Issue 1, 2016 • Vol 40 • No 1)
Read time: 6 minutes
In the event of an airplane crash, we would be horrified if a scientific, exhaustive and unbiased investigation was not conducted. Similarly, if told that a sunken ship failed to have a pre-planned evacuation strategy or an adequate number of lifeboats, we would be outraged at the criminal negligence by the ship’s captain. However, for some reason, pre-planned exit strategies and techniques to identify possible failures or understand the root cause of a failure after it occurs are extremely rare in business, despite the fact that the failure and underperformance rate for new products and companies is approximately 65 percent.1, 2
This omission is even more puzzling considering tools to mitigate or prevent failure require little effort or cost. Identifying useful techniques from other professions and applying them to a different discipline is a simple yet transformational act that can yield a higher ROI than any of the incremental optimizations performed by companies. Also, it is not as if these knowledge domain transfers do not work. Statistics, for example, began with the analysis of census data by governments many hundreds of years ago, evolving and improving until its adoption by every scientific discipline. For new product development, I propose the adoption of the following four practices:
- The failure mode and effects analysis
- The root cause analysis from the sciences
- The early warning system from economics and finance
- Pre-planned exit strategy
The Failure Mode and Effects Analysis
When planning the launch of a new product or a lead acquisition campaign, we usually conduct market research and data analysis, do some testing and maybe even form a cross-disciplinary team. In a few cases, the team might even identify a dozen or so things that could fail. However the challenge is to determine which of these causes is more important or likely to happen. The failure mode effects analysis (FMEA), which has been used for more than 60 years by engineers, is just such a tool since it identifies all the possible failure modes (e.g., the software is susceptible to malware) and then categorizes and rates each of these possible causes of failure on a one to 10 point scale based on their severity, likelihood to occur and ability to detect the failure with existing controls.
Once this process is done, users proceed to calculate the risk priority number (RPN) by multiplying all three categories: (severity) x (likelihood to occur) x (ability to detect). Then, it’s a simple matter of sorting RPNs in descending order and focusing on those with the highest scores. The next step centers around looking for ways to minimize the severity of a failure (e.g., a warning system), the likelihood it will occur (e.g., adding redundancy components) and/ or improving or adding detection methods.
Using Root Cause Analysis
When a business, product or campaign fails, a person or some external factor (e.g., the competition) is often blamed. If asked, management will state that was the root cause of the failure. The conundrum is that while those are possible immediate causes of failure, they are unlikely to be the root cause. When the Challenger and Columbia exploded, was NASA able to look up at the sky and determine that a particular person or part was to blame? It really is no different in the case of a complex business; whenever a significant failure or underperformance occurs, businesses need to utilize a deductive process that uncovers the immediate, intermediate and root causes of the failure. This deductive process used by millions of non-business professionals for decades makes use of many techniques, but it primarily relies on fault trees and asking why each cause occurred, until the underlying causes are identified. One thing to always keep in mind is that root causes tend to be knowledge (e.g., faulty research based solely on purchase intent questions) or organizational deficiencies (e.g., the sales department did not share key competitive promotional activities with the product manager).
The Early Warning System
Most product managers use dashboards that monitor key metrics and drivers such as the number of units sold, profitability, etc. These date points are all critical metrics but are something that economists call “lagging indicators.” Lagging indicators are final outcomes that have other key contributing actions or events that precede them. Leading indicators precede events that can provide the early warning window for timely corrective action. For example, if selling a business-to-business product with an eight- to 12-month sales cycle and you react only after a lagging indicator falls below forecast (e.g., orders booked), you would have wasted more than eight months. In this example, the key would be to identify crucial leading indicators, such as the ratio of appointments to presentations to quotes. As a result, you might identify that, although you are making the usual number of presentations, you are not getting the normal number of quote requests, which will probably result in fewer orders.
Economists and finance professionals have successfully used early warning systems for many decades, identifying key leading indicators and assigning an importance weight to each one before calculating a sum total score. The process involves identifying the key leading indicators that drive revenue, entering their forecasted and actual numbers, calculating their variance and assigning an importance weight to the lagging indicators. While the lagging indicators generate the weighted score, it’s the variance in the leading indicators you focus on for early remedial action.
The Pre-Planned Exit Strategy and Trigger
Unfortunately, as I mentioned at the beginning of this article, while business professionals may spend a lot of time planning the launch of a new product or business, they seldom include any pre-planned exit strategy and what should trigger its consideration.
When designing a pre-planned exit strategy, one key component is the exit trigger, a signal that you need to go beyond the usual remedial actions as illustrated in Figure 1.
Figure 1: Overview of Exit Strategies
When the exit trigger is activated, basic remedial action should have already taken place using the early warning system, including a root cause analysis of any failures. The following are some typical considerations in this process:
- Determine the probability of success of these additional remedial efforts
- Financial goals and target dates
- Market considerations (e.g., the entry of a superior technology)
- Identifying which market exit option is best suited for your situation
Although no one wants to fail, the question is more often if you want to go all-in on a new product with blinders on. After many remedial efforts have failed, do you go down in a blaze of “glory” or live to fight another battle under conditions where the odds of success are more favorable?
About the Author
Alexander Edsel is a faculty member teaching graduate and undergraduate classes at the University of Texas at Dallas and author of the book “Breaking Failure” by Financial Times Press, Pearson.
- Griffin and Paul Belliveau, Drivers of NPD Success. The 1997 PDMA Report (Chicago, Product Development & Management Association, 1997).
- National Federation of Independent Business Education Foundation, 1997.