It's an old engineering adage that you can't have it all. As the saying goes: "Faster, better, cheaper — pick two." The idea is that any system can optimize at most two parameters, to the detriment of the third.

But there's another way to look at "faster, better, cheaper." It's also a good rule of thumb regarding the characteristics that are important in a given economic environment. And here, you only get one choice.

In a period of hypergrowth, such as the dot-com boom of the late 1990s, "faster" ruled the roost. When I was CTO during the boom, if I gave my boss two choices -- "You can have it in six months, for $1 million, or in six weeks, for $10 million." -- he would invariably go for the latter. And he wasn't wrong to do that: In those hypergrowth times, the value of getting an offering to our customers faster outweighed the cost of doing so.

In a recession, it's all about "cheaper." Pretty much everyone reading this has been asked to do more with less. Eighty percent of the IT professionals I work with say they expect their budgets to be flat, or to decline, for the next year. And that leads to decisions such as "cut costs at any cost" — including decreasing services to the business units, or delaying much-needed upgrades and refreshes. Again, this is often the right decision for the times: If the choice is between laying off irreplaceable employees or delaying laptop upgrades for another year, the employees might be the better bet.

Finally, in a period of modest growth, the watchword is "better" — architect a solution that's priced right for the short term, but continues to deliver value in the long term. That's not the same as the fastest solution, or the cheapest solution — it may be faster or cheaper than some alternatives, but it's usually slower or more expensive than others.

This all sounds pretty obvious, right? Interestingly, it's not. Vendors often try to pitch their solutions as "faster" (or more powerful, or more scalable) when the market is demanding "cheaper" or "better". And IT departments tend to push harder for "better" solutions, because we tend to have an innate distrust of optimizing for extremes.

So what should folks be doing right now? As noted, most IT groups are predicting another year of flat or declining budgets, so on the surface that would argue for "cheaper." But hidden in that gloomy data are glimmerings of a change. Back when we first started asking the question (in January) a full 98% of IT departments said they expected budgets to be flat or declining. By June, the percentage that expected budgets to rise grew from 2% to 20%.