New Product Failure Rate is an Urban Legend


When you read articles about innovation and specifically about new product failure rates, often Harvard Business School professor Clayton Christensen is quoted. It is claimed that he has said that “there are over 30,000 new products introduced every year, and 95% fail”. If this were true, I think that innovation/ product development/ starting up a company would be the riskiest thing you can do in business. For existing companies, it would promote the idea that sticking to existing products is the best thing to do.

Luckily the quote proves to be a so-called “urban legend”. George Castellion & Stephen K. Markham have asked professor Christensen if he has stated this, which he denied. But it is not only Christensen that (in this case allegedly) claims that there are exorbitant failure rates. In their article “Myths about New Product Failure Rates” (2013) Castellion & Markham quote several sources that claim product failure rates are 80% or higher.

Definition of Product Failure

The first question is, what is a Product Failure? Is it when we don’t launch after all, or if we launch to find out that we cannot produce at the right quality? Or is it when we are not meeting the business case or a certain minimum revenue/ profit amount? The definition that you choose highly determines the success or failure rate of the course. For clarity’s sake, let’s stick to the definition of Castellion/ Markham.

New product failure rate is referred to as the percent of new products introduced to the market and then failing to meet the commercial objectives of the business unit that launched the product.

I believe this also leaves space for interpretation:

  • How do we know if the commercial objectives were realistic or not? If I set my expectations lower, my success rate will go up, so I can manipulate my success rate, right?
  • Isn’t there a big difference between meeting 5% of the objectives or 95%?
  • What is the timeline for these commercial objectives?
  • If we sell more, this could be perceived as a failure (though I guess they intend to say “to meet as a minimum the commercial objectives”)

They also make a clear distinction between Product Failure Rate and Idea Failure Rate.

Idea failure rates refer to the percent of ideas that enter the development process but are not launched as commercial products.

Is a high idea failure rate a bad thing?

To start with ideas, we should strive to have a high idea failure rate, which decreases the further you move into the development process. In older research, it was found that “roughly 6% of all official ideas and 14% of the promising ideas that reach the development phase become a commercial success”  (Kerka et al., 2009; Liberatore & Stylianou, 1995). So for ideas, ‘failure’ is not the right word to use. It is about the selection and validation process, and in this process, we move through the following stages

  1. idea
  2. interesting idea
  3. promising idea + adjustments
  4. good idea + adjustments/ pivots
  5. great idea + adjustments/ pivots

I have added the adjustments/pivots because an idea will/should evolve into a better concept based on customer/user feedback. If your idea funnel looks a bit like the below image, it means that you have a good selection/ validation process in place. You need many ideas to uncover a couple of successful ones, and for this reason, many companies are professionalizing their idea management process. However, this also implies that you need a high and constant inflow of potential quality ideas. It calls for the professionalization of your idea management.

You need proper idea management to ensure capturing the most successful ideas. Capturing successful ideas require sufficient inflow AND good selection mechanisms for ideas.

Innovation funnel

Real Product Failure Rates

So, product failure rates are about products that are launched in the market not meeting their commercial objectives. What then is the real, more realistic Product Failure Rate? In the article of Castellion/Markham, they looked into 19 peer-reviewed research studies in the period between 1945 and 2004. It turns out that the failure rates range between 30 and 49%, for more than a thousand business units across over ten industries.  The range can be explained because there are differences between industries.

A PDMA research shows that only 61% of the products launched were successful. The PDMA also shows that there are differences between various industries, product types, technology bases, market types, and sizes. Details can be found in the table below.

product success rates data

Also in the Food industry success rates are much higher than people might think according to this article.

Unfortunately, I have not been able to find more recent research. I am curious if a distinction in failure rates can be made across the various types of innovation: incremental innovation (Horizon 1), more innovative development (Horizon 2), and radical innovation (Horizon 3). I would expect that the failure rate increases the more innovative the solution is. However, on the upside, there also is the potential for a bigger return.

For startups, I have found the following overview (I haven’t checked their sources though).

startup failure rates

Strategyzer claims that we can use the success rates for Venture Capital as a proxy for success rates for companies.  They even say that the ratio might be more extreme because established companies will be more risk-averse and less innovative. I think you can only compare the Startup Failure Rates ), more innovative development (Horizon 2), and radical innovation (Horizon 3) with Venture Capital success rates.

Strategyzer return distribution venture capital

Source: The Invincible Company, Strategyzer

increasing product success rates by 30%?

So, we have established that the failure rate is around 40% and most likely higher for more risky types of innovation. Of course, now it is interesting to analyze what the common reasons are for failure and how to prevent them. And we have to be realistic, a 0% product failure rate can never be achieved. Markets, people, technologies, and society all change and influence the chances of success.

But imagine that you could improve your innovation excellence and hence reduce the risk of innovation? What if you could reduce your product failure rate or more positively stated increase the product success rate? What if you could reduce the time spent working on ideas that are not worth pursuing? What if you could learn to fail fast and cheaply? (As opposed to failing late and expensive which is the case when you have a product failure.) Can we set the target at 30%? So if your success rate currently is 60%, you want to achieve a 78% success rate? A 30% improvement in your operational excellence would make the entire company proud of the achievement. Why not try to achieve this in your innovation excellence and hence improve the product success rates?

Let’s go for a 30% improvement in your product success rates!

It can be done according to the PDMA research! It reveals that if they compare the outperformers with the rest, these best companies have an average success rate of 82% versus 53% for the rest.


In our next article, we will dive into the reasons for failure and share best practices that you can apply to improve your innovation excellence and product success rates. So stay tuned for our next article!


Wanneer u onze website bezoekt, kan deze informatie van bepaalde services via uw browser opslaan, meestal in de vorm van cookies. Hier kunt u uw privacyvoorkeuren wijzigen. Houd er rekening mee dat het blokkeren van sommige soorten cookies van invloed kan zijn op uw ervaring op onze website en de diensten die we aanbieden.