See WorseIsBetter
Time to market
Easily the top reason. Management often believes (whether they're right or wrong) that it's more important to get even a low-quality product to market as fast as humanly possible (or faster), and to do so typically on a shoestring budget, than it is to have a high-quality product that they believe would take longer and/or be more expensive.
Quality is relative
Different consumers have different needs; what may seem like an excellent product to a minority of people may not address the needs of the majority.
Timing is not right
I would suggest IT companies that spent a lot of effort in security concerns in early days of the WildWildWeb? would have gone bankcrupt, or much diminished in size. Look at the approach of IbmCorporation vs MicrosoftCorporation during the late eighties and early nineties. The former took a lot of effort in ensuring backward compatibility, reliability, etc etc.
Perceived quality may not match the real quality
This one cuts two ways: a product or feature may seem advantageous at first glance but prove to be impractical, or a poor quality product may have a reputation of being superior.
Poor business management
A product cannot succeed for very long if the manufacturer mishandles their business management. Various aspects of this include:
Poor reputation
Sometimes, a product gets a bad reputation which it simply cannot shake, even with the best marketing and advertising.
Higher quality can drive up the price
Consumers buy for cost reasons as well as for quality, and many are not willing to pay a premium price for a product, even if the less expensive product needs to be frequently replaced or repaired; this can also feed into the produce cycle issue below.
Not following industry standards
Entrenched competitiors
Slow replacement/upgrade cycle
If a product rarely needs to be replaced, repaired or improved, the manufacturer may do less repeat business than those whose product has a faster replacement or upgrade cycle, who are thus more profitable. This sort of thing tends to be self-reinforcing, as the faster production cycle tends to degrade product quality, forcing an even faster upgrade cycle (It should be noted that, for the most part, the forces which drive the upgrade/replacement cycle for OpenSource software are different from those in the commercial world, such that the rate of change does not necessarily affect the quality of the software either positively or negatively).
This can only occur under certain circumstances. First, this phenomenon can reinforce success, but cannot create it; the 'lesser' product must already be established and have a competitve advantage over the 'superior' one. Second, the actual perceived difference in quality must not be a significant factor in the purchase choice for most consumers. Finally, the cycle will eventually bottom out when quality drops below the threshhold of consumer acceptibility (though the pattern can take several cycles to reach that point, and usually other market factors will interrupt the pattern before it reaches the nadir; this is why the ShoeEventHorizon never occurs in the real world).
Quality occurs at a cost
Otherwise everyone would be doing it because higher quality provides higher satisfaction amongst workers who contributed to it.
One of the higher cost is cost of maintenance. People need to be continually trained/updated in best practices and it is costly. When a new technology comes along it is costly to design a "quality process" to fit the application of new technology and make the whole process (including legacy infrastructure maintenance) a "highquality process" -- dl
It's worth noting that one of the reasons for the gradual encroachment of foreign auto makers in the US car market is the insistence of said makers on turning out a quality product, while the US makers continue to downplay the importance of quality.
Buyers looking for dependable luxury almost reflexively look to Toyota/Lexus, Honda/Acura, Mercedes/Daimler, BMW, Audi, even VW over GM/Cadillac or Ford/Lincoln.
This may not be the case in a market where none of the players invests in quality, but once any player begins demonstrating consistently higher quality, eventually even FUD can be overcome and serious incursions into the market can be achieved.
The US once derided "cheap Japanese" products, injecting liberal dollops of FUD into discussions, but after real quality became the rule, eventually the critics had to acknowledge the viability of that approach.
If you intend to be in the game for the long haul, it would seem that quality pays large dividends.
But it's worth noting that when those "cheap Japanese" products first started to make incursions, it was definitely due to low price and not quality -- the latter was sorely lacking initially. (I remember with paradoxical fondness my rusting and breakdown-prone Datsun B210 of college days.) Which may illustrate that the "right" level of quality is a marketing decision that changes according to one's goals and market conditions.