Microservice is a fast growing and widely accepted software architecture characterized by collection of services that are intended to perform specific tasks. The decentralized arrangement provides greater flexibility and scalability. The intricate nature of these architecture makes it difficult to tackle down the root causes for a failure. The identification of causation metric within the microservice, among the various metrics is challenging. The proposed methodology localises the root cause metric by utilising the trace anomaly score collected from traces. Upon localizing the faulty service, the root causation metric within the service is identified. A sequence of procedures encompassing trace anomaly detection, suspicious microservice set mining and microservice ranking is utilized to determine the faulty service. Within the service, the root cause metric is identified by incorporating correlation analysis between anomaly score and each metric. The key data source for the experiment was traces, which proved valuable for the widespread application in recent approaches. The approach draws a connection where the change in the anomaly score is reflected in the underlying cause metric. The benchmark Train-Ticket microservice trace dataset was utilized for the experiment. The approach proves to be better when compared with the existing method that uses log data. It shows 32% increase in top-5 precision. The time efficiency exhibited was also superior to the existing method. The proposed work can efficiently help in mitigating the failures from the traces.