Does quality really matter anymore?

Many products now fail after just a few years, and we seem to be OK with that.

I have been thinking about a simple question: what has been the trend in consumer product quality over the past few years and decades for electronic-centric and electromechanical products?

Unfortunately, this very basic question has no definitive answer. As with most such queries, the best answer is “it depends.” In this case, it depends on what you mean by “quality” and its complement of “reliability,” as well as the specific classes of products you are asking about.

While there are formal definitions and metrics for quality and reliability, I am more interested in the average person’s understanding of what those two overlapping terms mean. Further, I am excluding software bugs that contributed to problems ranging from occasional crashes to outright non-functionality; instead, I am focused on hardware-based physical intermittent and outright failures.

I’ll start with one product group for which there are solid numbers: automobiles. In general, their quality has increased dramatically (with notable exceptions), largely due to the widespread use of components qualified to AEC-Qxxx standards, the application of ASIL safety levels, rigorous design-review assessments, the inclusion of circuit protection, and adherence to stringent requirements. This has happened despite the vastly increased complexity of vehicles, which now feature many sensors, processors, active circuitry, various functions, motors (a typical car has 40 or more small electric motors), and interfaces.

Of course, the inclusion of all these circuits and hardware also means that when something does go bad, it is far more costly to fix. It often must be done by the dealer or specialized repair shop; the days of being able to fix your car with the replacement part and a screwdriver and a wrench are long gone.]

What about other consumer products?

The situation changes when you look beyond cars and focus on other consumer items, ranging from lower-end products like network routers and various interface boxes, to items such as smart clocks, and larger items like large-screen TVs and home appliances.

My sense here is that their quality has declined, but it’s a very foggy situation. While I would like to base this assessment on hard data, there are challenges in doing so. There is very little substantive data available, except for household appliances, as shown in Figure 1, since there is no formal data collection scheme. Most people just toss the defunct product and move on.

The alternative is to use self-reported surveys among consumers, but these have large margins of error. Finally, there are numerous anecdotes, but these can give a skewed picture.

The problem with the reliability of small- to medium-sized products is that good designs require extra attention to component selection, design margins, and manufacturing processes. That’s tough to do when profit margins are strained and customers have been conditioned to have low expectations for these low-end devices.

What about repairability? First, you must be able to access the product, but many are designed to be closed at the factory and never reopened, ostensibly for user safety (“no user serviceable parts inside”), but also to reduce assembly and bill-of-materials (BOM) costs. It’s no secret that a dense PC board with a defective component is challenging to troubleshoot and repair, and a complete replacement PC board is generally not an option.

The directive to the design team is “just make sure it lasts three to five years,” and there’s no need to worry beyond that period. It’s a consumer mindset that is contrary to that of just a few decades ago.

It’s especially frustrating when you see products like flat-screen TVs being tossed in the trash. Has the owner decided to upgrade? Possibly so, but it is more likely that the screen went bad, it is too costly to fix, and you can upgrade to a better one for less than you paid for this one.

This is one of the unintended and unforeseen consequences of Moore’s Law coupled with high-volume, highly efficient manufacturing. Further, using reliable components is a necessary but not sufficient tactic for a reliable end product.

Examples demonstrate reality

I can provide at least a dozen personal cases where a modest home product “died” because of design shortcuts, often where the internal bulk-filtering capacitors in the DC-converter subsection degraded after only a few years. This fault mode is well known, and you can buy longer-life capacitors for a small premium, but why bother? It’s especially frustrating since using better capacitors does not require any change to the design, layout, assembly, or software; it’s just a bill-of-materials (BOM) upgrade at slight cost.

I’ve tried to push against this trend in my own small way by replacing those failed capacitors where possible. Sometimes they are accessible, but often they are not, assuming you can actually get into the product in the first place. The $15 line-operated clock of Figure 2 was tricky to open, the capacitors were hard to reach, and I had concerns about reassembling it safely with the AC circuitry and the low-voltage sections so close to each other.

In contrast, for the $25 digital-to-analog TV converter shown in Figure 3, which allows you to use an old analog CRT TV with digital OTA broadcasts, the AC side was on its own circuit board, with capacitors that are easy to see and access.

Despite the time and frustration involved, I regard it as a small victory against a huge wave of products being discarded due to reliably minor failures, when I succeed. Of course, average consumers won’t be inclined to try this.

Given the lack of credible data, my sense based on anecdotes, message boards, user surveys of dubious credibility, and some coarse data, it seems the quality among lower-to-middle range products has declined. It’s a sad contradiction to the “quality doesn’t cost, it saves money” movement and associated efforts of just a few decades ago that drove companies to significantly improve their designs, yields, and overall quality.

Why do I care about this? Besides the huge environmental-waste impact of all these discards, it casts a negative light on engineers and diminishes our credibility. While engineering was — and still is ­— an intense and rigorous disciple, these lower-quality designs radiate the contrary to the general public. While some individuals may appreciate the relentless improvements in products along with lower costs, it also diminishes their assessment of the engineers who make it happen.

References

Assessing Product Reliability, NIST
Product Reliability Data, Product Management World
Why don’t consumer products last as long as they used to?, Quora
The Most Reliable Appliance Brands for 2025, Yale Electric

Related EE World content

The winding path to the infrared remote control: part 1
The winding path to the infrared remote control: part 2
Does 32 kHz or 3.58 MHz mean anything to you? part 1
Does 32 kHz or 3.58 MHz mean anything to you? part 2
Confidently sourcing electronic chips on the open market
Collect and analyze production test data in the cloud