No one has accidentally made something ‘too’ reliable.

There are lots of conversations in design, manufacturing and production meetings where someone says something like …

… we don’t want to make our product too reliable …

There is often a lot of furious agreement and head nodding. Some say that if we make something too reliable, we will spend too much money during production in a way that won’t be worth it. Others will say that we don’t want our products to last too long, otherwise our customers won’t come back and buy more products, or upgrade to the next version.

Here is the problem with that idea. We know of thousands of companies that are now out of business because their products were unreliable. But we don’t know of too many companies (if any) that are out of business because their products were too reliable and rarely failed.

Toyota is the largest seller of vehicles in the world and is also famous for these vehicles being perhaps the most reliable. Some people might have heard of the ‘Yugo,’ which is a car deliberately made without any consideration for reliability and quality to reduce their purchase costs. And the company that made them no longer exists.

So why do we keep hearing concerns that we might be ‘dangerously’ close to making something too reliable?

Humans often look for facts to support their biases and emotions

Lots of designers, engineers and manufacturers don’t want to take reliability and quality seriously. This often starts with them being taught about reliability and quality activities in terrible ways, by terrible teachers. There is often a preponderance of statistics, suffocation by standards, and a vicious focus on the trivial while ignoring the more useful focus on the vital few quality and reliability challenges.

So it stands to reason that people are not motivated to do reliability and quality stuff. And that means we often try and validate this aversion by imagining fake barriers and presenting them as facts.

This is especially true when we are culturally led to believe that all we need to do to make something reliable and high quality is to simply ‘do our jobs.’ There is an element of truth to that. Cutting corners never ends well when it comes to design, production and manufacturing. And as most people tend to not doubt their own abilities, we can often walk around as if all we need to do is employ our prodigious talents in everything we do … and we’ll be fine!

But again, if this was entirely true, there wouldn’t be those thousands of companies and manufacturers who have gone out of business because their product failed too often in the hands of their users and customers.

But customers buy new things even when their old ones haven’t failed!

"Smartphones are ‘approaching washing machine territory’, meaning they are generally only replaced when they are broken."

-Ben Wood, CCS Insight (2024)

People often think reliability is all about how long something will last. But it is also about how many things will fail in a period of usage. For example, ‘Smart Phone A’ might last (on average) for 6 years before it fails. ‘Smart Phone B’ might last (on average) 8 years before it fails. So ‘Smart Phone B’ is more reliable … yes?

Well, even though ‘Smart Phone B’ lasts longer (on average) than ‘Smart Phone A,’ it turns out that it fails more often in the first three years of use. Around 7 % of ‘Smart Phone B’ devices have failed by 3 years, while around 2 % of ‘Smart Phone A’ devices have. So ‘Smart Phone A’ is more reliable … from this perspective.

So even though most customers might (for example) use their smart phones for three years before they upgrade, along with the fact that both ‘Smart Phones A and B’ will on average last for at least twice as long as this 3-year period, failure is still a huge problem. Sometimes you can only tolerate 5 % of your products failing during warranty or useful life periods. And with more than three times as many ‘Smart Phone B’ devices failing in that 3-year period as ‘Smart Phone A’ devices, customers will very quickly form an opinion on which one they will buy ‘next time.’

So you can never wish reliability away.

And as Ben Wood identifies above, the period in which people are holding onto their smart phones is increasing. So if you aren’t taking reliability seriously, you aren’t taking your own job security seriously!

But customers don’t really value reliability when they buy things!

Let’s go back to vehicles. The chart below shows the average age of vehicles in the United States in the 21st century.

S&P Global Mobility, U.S. average age of light vehicles as of January 1 for each year shown

This means people are waiting longer to replace their vehicles than ever before. In the 1950s, the typical vehicle warranty was 3 months. Now vehicles are being offered with 10-year warranties. And the market is responding accordingly. Which means people care about how long things last.

So if you hold onto your Toyota vehicle for 15 or more years and have had a good experience with it, we know that you are more likely to purchase another Toyota and so on. Which is why Toyota is the worlds largest vehicle manufacturer (it’s worth repeating).

So why doesn’t quality and reliability happen more often than it does?

It starts with the people who teach reliability and quality techniques.

Teaching is not just about teaching people what to do and how to do it. It must start with why they do things. Many successful companies understand that reliability and quality embedded into design, manufacturing and production actually saves money from the very start.

What? Don’t reliability and quality activities incur additional costs?

Yes they do. But when you start off with the simplest ones up front, you have the ability to implement really cheap and smart changes to your product or manufacturing lines before you even think about locking down their first designs. Addressing problems early saves huge amounts of time and money. A simple change up front can replace a months long ‘crisis’ just before you hoped to launch your product.

And by thinking about reliability and quality from the very start, you keep track of your progress. So you never ‘accidentally’ fall into the chasm of ‘too reliable!’ Which of course … doesn’t exist.

So the next time you hear someone bemoaning the perils of making something too reliable, ask yourself if this is based on facts or emotions. If people are saying this because they don’t want to take reliability and quality seriously, then you have a problem. If you are the leader of this group of people, then you need to fix this.

Otherwise you will be consigned to history in the same way the ‘Yugo’ was when it was voted the worst car ever made in the 1980s.

---

Read more on Decision Making

Read more on Leadership

Read more on Business Management

Read more on Reliability Engineering