A rough data structure was derived from the primary database (comprised of the information from Berg and the NAIP which would enable the study of specific physical aspects of the imprints, including type, paper, typography, illustration, and binding. However, in order to insure that this digital skeleton was going to bear the load about to be placed on it, copies of the preliminary versions were sent to experts around the world.

Over 100 special collections librarians, bibliographical scholars, and book arts historians were contacted and asked to review the initial list of characteristics the project was planning to collect. If they found aspects missing from the list that they felt were important things to consider, they were asked to add these with explanatory notes. The response was overwhelming and their generous suggestions kept the project on target and actually increased the degree of detail on which it was based without sacrificing the overall scholarly goal.

The resulting design is among the most comprehensive bibliographical database structures in the world. For each imprint over 200 characteristics are considered. Not all of these fields apply to each imprint (binding data is not relevant to most ephemera, for example), and estimating the amount of time required to fully evaluate a piece can be tricky. Some imprints can be surveyed in a matter of minutes while others have taken as long as seven days.