Presumably, he is not going to personally attempt to investigate all hundreds of millions of data points. Instead, he'll devote whatever amount of time he feels is required per point, and an extra couple of seconds on putting a pin at either end of the process. Or, alternatively, he will centre in a specific subsection of space, and use the pins to devise patterns, groupings or other metastructures in a visual way, because it sometimes does pay to operate in non-virtual environments.
ETA: that is, if it wasn't a con, which it turns it was. But I stand by my general statement, as someone whose best work tends to happen at the pen-and-paper phase of the development.
Grey Wolf