Failures cluster at the beginning and the end, and the middle is quiet

These are listed in the order worth acting on, which with failure rate over a product life is not the order they are usually presented in.

The three regions of the curve

Early failures come from manufacturing defects: a bad solder joint, a mis-assembled part, a component outside tolerance. These appear quickly under normal use and produce a high failure rate that falls away over the first weeks or months. The middle period has a low and roughly constant rate, driven by random events such as power surges and accidental damage.

Wear-out failures then rise as bearings, seals, capacitors and mechanisms reach the end of their design lives. The resulting shape is why the pattern is described as a bathtub, and it applies across product categories with surprising consistency.

Burn-in and why it exists

Manufacturers of critical equipment run products for a period before shipping specifically to move early failures inside the factory. Consumer products rarely receive this treatment, which is why the first month of ownership carries the highest failure rate.

After the warranty ends, this is the practical justification for testing everything thoroughly in the first days, while return rather than repair is available. It also explains why a product that has worked for six months is statistically much safer than a new one. Refurbished units have often passed through this early region already, which is a genuine and underappreciated point in their favour.

Reading a warranty against the curve

A warranty covering only the early region protects the manufacturer against defect claims while excluding the wear-out region entirely. Extending a warranty into the flat middle region is cheap for the seller, because the failure rate there is low.

Extended cover that ends before wear-out begins is therefore priced against a period when little happens. A manufacturer offering long cover on a specific component, such as a motor or a compressor, is making a claim about that component's wear-out point. Comparing the warranty length against the expected wear-out age for the category is a quick and revealing exercise.

What changes the shape

Better quality control shortens and lowers the early hump, which is what a reputation for reliability usually reflects. Better component selection and derating push the wear-out region further right, extending the flat middle. Harsh environments raise the flat section, because random external events become more frequent.

Maintenance affects the wear-out region directly, since replacing consumables before they fail resets part of the curve. Design changes mid-production can reset the early hump, which is why revised versions of a model sometimes behave very differently.

What a fault tells you about the future

A fault in the first weeks usually means a defective unit rather than a poor design, and replacement is the appropriate response. A fault in the flat middle is often a one-off, and repair is usually worthwhile because the rest of the product is not near wear-out.

After the warranty ends, a fault in the wear-out region is a signal that other components are also near the end, which changes the repair calculation. This is the honest reasoning behind the intuition that old machines start failing repeatedly once they begin. It also means a single expensive repair on an old product is often the start of a sequence rather than a solution.

Using this when buying

Test thoroughly and immediately, because the early region is where returns are easy and defects are most likely. Discount short warranties appropriately, since a term ending before wear-out is a limited promise regardless of its wording.

Look for reliability survey data covering several years, which is the only way to see where wear-out begins in a category. Treat a very cheap product with a long warranty as an invitation to read the exclusions carefully. This site does not gather failure statistics, so surveys published by consumer organisations and trade sources are what fill this in.

Everything above, in order of what to do first

1. The three regions of the curve. Early failures come from manufacturing defects: a bad solder joint, a mis-assembled part, a component outside tolerance.
2. Burn-in and why it exists. Manufacturers of critical equipment run products for a period before shipping specifically to move early failures inside the factory.
3. Reading a warranty against the curve. A warranty covering only the early region protects the manufacturer against defect claims while excluding the wear-out region entirely.
4. What changes the shape. Better quality control shortens and lowers the early hump, which is what a reputation for reliability usually reflects.
5. What a fault tells you about the future. A fault in the first weeks usually means a defective unit rather than a poor design, and replacement is the appropriate response.
6. Using this when buying. Test thoroughly and immediately, because the early region is where returns are easy and defects are most likely.

The takeaway

Test hard at the start, relax through the middle, and treat the first wear-out failure as information about everything else.

Anything you cannot buy a spare part for is a rental with a long term.