Addressing Delays in Agent Call Execution During Testing

I have a single request that went through during testing this morning and so far it's been in queue for 15 minutes. The goal is to have the agent call the submitter within the delay that I specify in the flow. Having a huge delay before the agent calls defeats the purpose all together. Can you guys provide some insight as to why there is such a big delay in execution and is it something you are working on?