The term latency, in regards to data, refers to the time it takes from data being captured to it being available for reporting any analysis.
Splunk is able to capture and analyze data in real time (that is, in under one second, often in hundredths or tenths of a second) when deployed on appropriately sized hardware. For example, if a Splunk alert triggers, it can run a script to immediately execute a server shut down. If a denial-of-service attack (a cyber attack that can dramatically hurt an e-commerce company's bottom line) is taking place, Splunk can be used to figure out what is happening in real time.
If data is not available in real time, Splunk can monitor and capture when the data is made available, run data capture services on a scheduled basis pending the nature of the source data and system, or just listen on a server port awaiting for and accepting data once it arrives.