Delay Analysis Methods
As-Planned vs. As-Built
A direct comparison of planned dates against actual dates -- the simplest retrospective delay technique, but one that can't by itself distinguish real delay from an activity's slippage simply being absorbed by its own float.
Definition
As-planned vs. as-built lines up an activity's baseline dates against what actually happened and reports the variance. It's the most accessible starting point for understanding a project's history, but on its own it conflates two very different outcomes: an activity finishing late while it still had float remaining (which didn't delay the project at all) and an activity finishing late after consuming all its float and pushing the finish date (which did). Reading the variance correctly requires checking each slipped activity's own float, not just noting that its actual date differs from its planned one.
Example
An activity with 15 days of float finishes 8 days late -- still 7 days of float remaining, no project delay. A separate activity with only 3 days of float also finishes late, by 10 days -- consuming all 3 days of float and pushing the project finish back by the remaining 7, a real delay.