Skip to main content
Vice President, Asia Pacific
Tags: General , Hardware
Date: May 10, 2017

Your Product is Going to Fail: The Bathtub Curve

Your Product is Going to Fail: The Bathtub Curve

Some of us are early adopters of technology. That long line of people camping overnight to get the latest phones? They're definitely early adopters. You can't miss early adopters—even when they're not in lines. Because they're also the first ones to complain when problems crop up.

And problems always crop up.

I'm not an avid video gamer, but I can't help noticing headlines about Nintendo's latest console, the Switch. The headlines tell me two things: first, the new hardware is selling like hotcakes, and second, that there are issues with the new console—particularly with its controllers losing connectivity.

My bet? In a year or two, you'll be less likely to hear about problems with the hardware.

Failure to compute: Why we fall into the first 90 days of the bathtub curve

Reliability engineers know this kind of failure as the beginnings of what's known as the bathtub curve—the life cycle curve that graphically illustrates the rate of failure of a product over time.

The highest failure rate of any new product typically occurs in the first 90 days, during what's called the "burn in" period. Over time, problems go away as manufacturers iron out lingering issues, and that's where you see the fewest problems—at the bottom of the curve. The end of the curve shows the incidence of bugs increasing over time as the product ages and reaches the end of its life.

This illustration looks like a bathtub, hence the name.

bathtub-graph.png

While all companies go through testing cycles to ensure the quality and reliability of their products are sound before they ship, problems show up nonetheless. The consumer market is a good indicator of this, as many issues tend to crop up, and demonstrate that you can't really put a stopper to the bathtub curve.

The problem—let's face it—is us, the early adopters. We're the ones who buy fast, and help vendors fail-test products. We're giving vendors permission to fail. We're telling them it's OK. We forgive you.

Even venerable companies like Apple that have traditionally scored high customer satisfaction numbers have a history of releasing revision A products with high incidents of failure, from the very first scratch-prone iPod Nano to iPhone 4 "antennagate" to iPhone 6 "bendgate" and many others.

You can't escape the bathtub curve, but you can choose where you land

You can (and absolutely should) do everything in your power to minimize face time with issues that accompany new hardware. Why stick your neck out for bragging rights to new hardware?

As decisionmakers for your organization's technology purchase, succumbing to the newest bauble can potentially wreak havoc on your life. If you've any doubt, just look at the story of Peak Hosting, a company that filed for bankruptcy after spending a fortune upgrading their systems to new and buggy hardware, losing their biggest customer, and facing a lawsuit as a result.

Let's pause there for a moment.

Imagine what kind of story Peak Hosting was sold on when they were considering the new technology. They were likely told: You'll be able to serve up content faster. Bottlenecks will be a thing of the past. And your customers will be delighted that you're investing in new technologies to better serve them.

Surely, they weren't told to read between the lines: We're selling you what looks like the best solution on the market, but you just have to look out for some issues that may surface. Any problems you experience, we'd be happy to fix them for you.

Except things didn't pan out that way.

Peak Hosting's big-named vendor didn't cop to the software bugs until the whole fiasco was over, by which time it was too late to save Peak Hosting from bankruptcy.

From my own experience shipping products, we don't always recommend the latest and [seemingly] the greatest to our own customers. It may seem counterintuitive, but our approach is: if you're looking to upgrade your IT infrastructure, make sure you use equipment that's been battle-tested.

We should all embrace the future through new technologies. Being the first to glean insights and benefits from a new solution is a tempting proposition. The hits can be great, but so can the misses—as the bathtub curve so clearly illustrates.

Go ahead and be an early adopter in your personal life. But in the corporate arena, it pays to be at the bottom of this particular curve.

Questions? Comments?

Talk to our team of expert engineers, product managers, and technicians by emailing us at experts@curvature.com

Share this