The delays and the capabilities meet in specific places. Reading the map from left to right gives the structure of the rest of this course.
Design delays meet pattern finding and prediction. If past trials show which protocol features tend to force amendments, that knowledge can be applied before a new protocol is finalized, and simulated outcomes can reveal an unworkable design before it is committed to. This is the subject of the next chapter.
Recruitment delays meet pattern finding and prediction as well, but on a different kind of data: health records. Software can scan large record collections for people who appear to meet the eligibility rules, and rank them by how well they match, so that sites spend their screening effort where it is most likely to succeed.
Data collection and retention delays meet prediction and automation. Predicting which participants are likely to disengage allows support to be targeted rather than uniform, and automating routine data checks shortens the gap between a problem appearing and someone noticing it.
Analysis and reporting delays meet automation most directly. The repetitive parts of analysis, and the drafting and consistency checking of long documents, are exactly the kind of rule-based work that software handles well.
The mapping is not a claim that each delay disappears. It is a claim about where effort is best spent, and it sets up the remaining chapters, each of which takes one row of this map and looks at it closely.