When one considers the various steps of the numerous shuffles, one has to, in order to be efficient, calculate a number of factors. There are also several thinks to take into account, which can not be so easily factored. Lets start with an example, first. Take the oh so common JC shuffle. The ore is purchased for 42 gold/stack. For this example I will list the numbers of my test last night. 48 stacks of ore were purchased at this price, and a tip of 150g was made during the whole course of a full shuffle, leading to a cost of 2450g, roughly (After factoring in heartblossom and jeweler settings). Now that the "Lazy Wes" shuffle has been nerfed (Referencing my good friend Wes of Cappedbycata.com, who before 4.2 had Obsidium ore buy able at 19gold a stack...) and you can no longer efficiently vendor all gems save carnelians for 5g, one has to carry the shuffle even further, in order to make profit, unless of course you are able to sell or dump all your other gems. Thus we bringing in the crafting and disenchanting factor. For my test I crafted everything into jewelry, and had a friend in a level 25 guild disenchant it. This left me with 39 carnelians, 40 zephryite, 10 stacks of dust, 25 GCE, one heavenly shard (DE'd a few blue rings, oops) and 7 blue pieces of jewelry. At current prices, this left me with 4450 gold in items. Since I prospected the ore while in dungeons, leveling my DK, I won't factor in that time. Multitasking FTW, eh? So for getting my items DE'd/crafted/bought/sold, etc, I factor in 45 minutes. Not terrible, around 2500 gold an hour, minimum. But wait! Now to decide what to do with the enchanting materials.
This leads into my second factoring portion. What to do know? How does one decide what to do? You have two main options, to sell the materials, or to make enchants and sell. Although while making the enchants will lead to a higher profit margin, there are several things to remember. One, what are faster to move? If you can continually move the raw materials, and get the ore at a price which is cheap enough to make it worth your while, you might be better off only taking the shuffle to this end, to prevent a bottle neck. After all, how many enchant bracers-critical strike can you move in a day? The undermine journal is a great place to figure out how often certain things sell.
An even more important factor is actually being able to sell things. If you can get rid of all your stock in a few hours as enchants, for slightly less profit, it is often worth it to do so. Example, if it costs me 107 gold to make a certain enchant (using the price I could get for mats) and I can sell it for 100g, but it sells right away I'd do that. I already profited by getting the materials for half price (2450 in costs for 4450 of materials) I can move stock faster, and I have an extra market to move stock in!
Thus, in a perfect world, I can be selling both the materials I don't need, and the enchants, and increasing profit for every step I carry the shuffle. But in this practical world, sometimes its better to move stock in several markets (DIVERSIFY!) and move it quickly, to prevent bottlenecks, then to be sitting on lots, moving it far more slowly for a very little more profit.
But that is only in one market. Enchanting materials are, at least on my server, fairly stable, as are the profit margins. Thus you can learn how to move stock. But lets look at another market, belt buckles and red gems. These are polar opposites, the prices vary wildly within a certain range, but sell at a fairly constant rate. The key to remember here is that if you aren't supplying these markets, someone else is. I'd have to disagree with Cold of Cold's gold factory, where he states that he is preferring to hold onto his inferno rubies until his competition sells out. The problem is this, unless his stock is limited, he can't produce more, or there is a bottle neck, such as no one else creating inferno rubies, that this sell out will never come. As a fairly hardcore raider, I know that gear enters the game at a fairly constant pace. Every 2-3 weeks I will be getting a new tier piece, and I will probably be getting 1-2 other pieces a week, either PVE or PVP gear. Thus I will be using up ~4-6 meta gems (tank/DPS/PVP sets) and 15-20 inferno rubies over the next few months, as I gear up various sets, replace old 379 with 391 gear, replace boss drops with tier gear, or vis-versa. There will, of course have been a vary fast burst demand for gems, as people got their belts and new pvp gear. But that is pretty much over. Yes the market is more active now, but there probably won't come a time where all the rubies are sold out. Thus, try to reset prices where practical, try and get as high a profit margin as you can, but don't stop supplying or creating these items. If you don't supply us raiders, someone else will!