For some reason, as time goes on it bothers me more and more that the measure of a project's success is whether it's "on-time and under budget." Seems so outdated and uneducated.
In The Power of Product Platforms they quote a statistic that, paraphrased, states that:
Projects that deliver the most value (ROI) are the most late and the most over budget.
You can rephrase in a number of ways, but I always boil it down to: "if it's worth doing, it's going to be hard."
There are a few reasons I hate the "on-time, under budget" mentality, mainly that it rewards the wrong behavior (as happens any time metrics/rewards are not aligned with desired outcomes).
Instead, all projects should always have a Net Present Value associated with them that's calculated on an ongoing basis, updated as new data (costs or project revenue/benefit) is uncovered. Guess what, when a project's NPV goes below $0, the project is canned.