Assessing the creditworthiness of prospective business partners is the first step in conducting trade. Traditionally the creditworthiness of partners was assessed using transactional methods, methods based on close observation of the other party and the heavy use of mostly subjective, soft information. During recent decades, however, these relational methods were largely replaced with transactional methods relying almost exclusively on objective financial data, otherwise known as hard data. A considerable portion of firms involved in business to business trade are now small companies that do not to have any reliable or comparable financial information. In the absence of such information, transactional methods of assessing business creditworthiness have very limited practical value. To provide a remedy, this study proposes using non-financial information available from the Web and thus a return to the more transactional methods. We identify sources of credit-related information available from a typical business to business exchange and test if such information can predict firm creditworthiness. We conduct a study with a group of online businesses on a major B2B exchange and empirically show that a number of non-financial factors can in fact predict online businesses’ level of creditworthiness.