In order to track the loan lost-linking customers, we analyzed their historical daily consumption transaction network records (DCTNR), which include bank card transaction records, third-party payment transaction records, and network trading system order details records. We extracted the transaction date, time and address information from their daily consumption transaction path, analyzed the key factors affecting the tracking work, and constructed loan lost-linking customer path correlated index model which is applied to quantify the correlation between the initial search address and other addresses. In addition, we also establish loan customer daily consumption transaction network based on big data environment, propose the network sorting rules and searching rules, and construct the network sorting search algorithm to track loan lost-linking customers in different address types. In the case study, we analyzed the historical DCTNR data of a Chinese bank’s loan lost-linking customer, and applied loan lost-linking customer path correlated index model and network sorting search algorithm to track him in big data environment. The results represent that the method can achieve the purpose of tracking, and the tracking time and cost can be reduced by using network sorting rules and searching rules. It is of great practical significance and scientific guiding significance for banks, financial institutions and major financial platforms to apply big data, artificial intelligence and other information technologies to track loan lost-linking customers and recover economic losses.