One of the greatest benefits of the agile software methodology is that it reduces the feedback cycle. At a macro level this happens in the form of iterations. Weekly or biweekly cycles of work that gather formal feedback from the client on the work completed to date. These reduced feedback cycles are then taken another level further through their use in Continuous Integration build pipelines. As part of the build pipeline faster tests are run earlier to give instant feedback when a defect is found in the software system's build. This early feedback from the Continuous Integration system allows corrections to be made sooner, closer to when the problem was introduced and when the team has the most context on the problem area.
This mantra is then continued further in Continuous Delivery. One of the points Jez Humble and Dave Farley raise in their book is that if something is painful to do, do it early and do it often. Doing it early and doing it often is a direct approach to reducing the feedback cycle and getting feedback earlier. By deploying to production earlier and often, even if it is a fake production environment, allows you to refine the process as your failures happen sooner and probably more often. Eventually, after continuous refinement, the process will have been tested a large number of times and had all defects ironed out.
Reducing the feedback cycle is a great way to tackle a difficult problem. When working in an environment where expectations are constantly changing and there are a large number of stakeholders with differing opinions. Most problems in this situation can be solved by reducing the cycle time between when feedback is obtained. This doesn't mean that you will instantly produce flawless results on the first showcase because you've given yourself less time, it simply means that those incorrect approaches can be corrected earlier when changing direction is easier and people are more comfortable with throwing away all the work they have done and starting again.
This mantra of reducing cycle times is taken to a micro level through tools such as Continuous Tests and custom setups with Guard. These setups provide feedback to the developer every time they save the file. Informing them that the changes to the code they have made have either made the tests pass or made a few others fail.