There was a glitch in one of the backend servers which degraded performance of processing requests, this has since been resolved.
Been reviewing the architecture in the past few days and putting some optimisations to API processing with additional server resources to handle future loads / improve existing performance.