The turnaround is the thing that changed how we work. We scope a test, it runs, and the report is with us the same day, with reproduction steps our own engineers can follow rather than a list of CVEs to go and investigate. Every finding is something the agent actually managed to demonstrate, so the remediation queue is short and real instead of long and theoretical. Support has been excellent. We have only needed them a handful of times, but each time the reply came back faster than almost any other vendor we deal with, and the issue was sorted on the same thread rather than bounced around. I also like how much control you keep. You set the scope and the rules before anything runs, and you can have the agent check in before it does anything intrusive or let it run fully autonomously depending on the environment. Being able to choose that per test means we can point it at things we would not have been comfortable handing to a fully automated tool.