Unfortunately, I don't think it's possible for your basic question to be answered. There is simply not enough information on women's buying habits, particularly from before the twentieth century, and even when we have an individual woman's account books, we can't tell the exact garment that she was buying. (Barbara Johnson, an eighteenth century Englishwoman, created an album with samples of fabric she purchased, typically with the price and what she had it made into - and even with her thorough mindset, she neglected to mention anything about her underclothes.)

There are also some important distinctions between different parts of this timeframe. For instance, the long-waisted stays of the 1780s simply couldn't be worn under a high-waisted gown of 15-20 years later, while the corset styles of the late nineteenth century could fit under an outfit from 1905 or 1910, even if they wouldn't look just right. Also, the concerns of a woman living in a period where stays are generally made by a skilled tradesman would differ from the concerns of a woman who could sew a corset at home or a woman who had corsets available cheaply and in standard sizes at the local dry goods store.

#The province of the skilled staymaker (1680s-1790s)

In the mid-seventeenth century, it was the norm for gown bodices to be made with their own stiffening - see this example in the Fashion Museum in Bath, England, which was patterned by Norah Waugh, Janet Arnold, and the authors of the more recent Seventeenth Century Women's Dress Patterns. By the end of the century, a more loose-fitting gown was becoming very fashionable and popular - the mantua - which required a separate boned underbodice: the stays. Dressmaking and tailoring had been reserved for men by guild statutes, but with the advent of the loose mantua, seamstresses were able to fight for and claim the right to make this one article of outerwear. The male tailors, however, kept the rights to the more prestigious, more skilled jobs of making men's clothes, women's court gowns, and stays. (Fabricating Women: The Seamstresses of Old Regime France by Clare Haru Crowston and Dress, Culture, and Commerce: The English Clothing Trade before the Factory, 1660-1800 by Beverly Lemire go into this in depth, if you're curious.)

So at this time, the price of stays would reflect both the cost of materials (whalebone of varying widths and qualities could cost from 15 to 56 shillings per dozen, and several dozen would be needed for every set of stays; on the other hand, lengths of packthread, which could also be used to stiffen stays, cost about 8 shillings per dozen - they'd also require several layers of linen canvas, lining, pasteboard, binding, and often a finer exterior fabric) as well as the experience of the maker and his apprentices in cutting and stitching, not to mention fitting. Patterns of Fashion 5 features a lengthy explanation of the complex geometry and skill required for this trade that makes it clear that staymakers earned their fees. Stays were valuable enough (costing £1-£2 in the early eighteenth century, as much as a maid's yearly wage) to be a target for thieves, whether stealing them from a shop or a home or actually stripping a woman or child to get at them.

As a result of their value, it was not uncommon to mend them at home or send them out to be remodeled. While some wealthy women might buy multiple pairs a year, the vast majority did what they could with the ones they already had. In many cases, "the ones they already had" may have been secondhand in the first place! Staymakers often advertised that their wares were in the "present taste" or "newest taste", which shows that their customers had a developed sense of how well their foundation garments measured up to the current fashions.

A time of change (1790s-1810)

This will necessarily be a short section, as few people have studied corsetry in this period and my understanding of it is largely based on primary source research of my own, mostly for my master's thesis.

Fashion in the 1780s and early 1790s was inclining more to the "natural" in many ways, which, when it came to the body itself, meant a less obviously altered shape and a bust that was allowed to curve. The staymaking trade had no clear way to conform to this new fashion: for a century, it had used the same techniques to produce a conical silhouette, and these techniques could not be easily adapted for this purpose.

Corsets of this period typically mimicked the overall shape of earlier stays, but were much lighter, with only two layers total and spaced-apart whalebones. In some cases, these were just light enough to support the bust without flattening it, but it's coming clear as we look into unprepossessing stays tucked away in small collections that many women seem to have used garments shaped rather like the earlier stays, but cut to go under the bust. Experimentation by myself and Mackenzie Anderson Sholtz of Fig Leaf Patterns has shown that this style of corset works very well with the shift of the period to give support and create the look that we see in fashion plates and portraits. There's a pattern for a corset like this in the Historic Cherry Hill Collections on my blog.

It would have been relatively easy for stays to be altered to take on this new shape, and it's likely that many women went that route when it became clear that their stays would be otherwise useless. The extant stays of this type are very simply made, and most likely cost significantly less than the earlier stays, though, so it's quite possible that women purchased them new or made them at home.

A middle-class dream (1810-1860)

After this, of course, came the unboned white cotton "Regency corset", fitted with gussets to make curves for the breasts and hips. These were still made and sold professionally by people - now often women, who were even seen as the most appropriate people to do this work, as it still required being intimate with a customer's body but no longer required the same finger/hand/arm strength - but could also be made at home. One of the biggest bars to home-sewing of stays had been the difficulty of dealing with whalebone, but with corsets no longer requiring row after row of boning, any woman could, in theory, produce her own foundation garment. As the years went by and stiffening became more important again, many found that row after row of cording did the trick just as well.

While commercially-made corsets in the later 1840s and 1850s would change construction quite a bit, moving toward shaped pieces rather than gussets, it seems very likely that a lot of the heavily corded/quilted gusseted corsets in collections were used in that time period as well. Women would have had patterns that worked for them, and the knowledge to fit corsets made with those patterns to any changes in their bodies. What corsets did to the body during this period also didn't change as much as people think - most of the alterations in the fashionable silhouette were the result of alterations to the outer gown rather than the corset itself. The main change was the flattening of the bust in the 1830s-40s, and that could be done fairly easily at home as well.

The Industrial Revolution is here (1860-on)

In the second half of the nineteenth century, ready-made clothing was becoming more and more available - by the 1870s, a woman could walk into a store (or a set of stores) and purchase everything she needed for her entire outfit, from the skin out. They often didn't - dressmaking was still a tenable profession, and many things were still often made at home to save money and/or to be industrious - but it was possible. Professional corsetieres still existed, but the Worcester Corset Co. and R. & W. H. Symington were producing corsets on an assembly line in multiple styles and sizes.

As a result of exploitative capitalism and economies of scale, corsets could be sold at rock bottom prices, accessible to all women (even the ones exploited to make them). In the US, they could be found as low as $1 or 50c - roughly a day's wage to a maid, or less. That's a huge difference from the eighteenth century! It's clear from photographs around the turn of the century that all women didn't jump on the latest trends in silhouette, but that would have more to do with their relationships to cutting-edge fashion than their pocketbooks.

Something else to think about (reenacting/costuming)

This is a bit tangential, but it's important to keep in mind, when wondering "would they really have updated that frequently?", that women of the past would have been wearing these clothes all day, every day - and wearing them out accordingly. They also would be living the slow path through the years. While it's extremely annoying for us today to need an 1860s corset, and 1870s corset, and an 1880s corset ... they would have expected to go through at least three corsets in that time, and probably several more. For most women, the changing torso style wasn't the main impetus, but as they replaced their worn-out corsets, they'd end up with the subtly more up-to-date ones.

The same goes for things like hoops and bustles, which weren't hard to break. If you popped enough steels, well, you bought a new one, which would necessarily be in the new shape.