The way I interpret what the OP is asking for is that the customer can visit a page multiple times before a order is placed. An accumulating snapshot fact table has to be more deterministic in the workflow it is capturing. Quite simply, you could maintain the different events as differing fact tables. From there it is a simple matter of comparing aggregates in a ratio like fashion.

Between dates d1 and d2,

Entity A had x visitors and y purchases. Ratio --> y/x. You do this for all Entities (storefronts?) You dont exactly need to link the path for every user to get some sort of higher level conversion measure.