Abstract: Systems and methods for expense input utilities include providing an image of an expense-related document for display with an expense form. A selection of a portion of the image and a selection of an input field of the expense form may be received. Optical character recognition may be performed on the selected portion of the image to identify a text string and the selected input field may be populated with the identified text string. Metadata associated with the input field may be associated with the text string and the expense form may be converted into structured data that includes the metadata and the text string.