The only improvement I see is intercepting calls in a mobile application without injecting additional code. Currently, when instrumenting or monitoring a mobile application, it's necessary to release updates, prompt users to upgrade, and inject code into the backend. It would be beneficial if there were a method to intercept these calls without requiring users to download and upgrade the application. However, I'm unsure about the feasibility of such an approach. 

Additionally, there's potential for enhancing the pricing structure, particularly as scaling up can become expensive. Given the continual growth of the mobile landscape, revisiting the pricing model could be beneficial. 

Though I haven't explored AppDynamics Mobile Real-User Monitoring a lot, considering the way the setup of the product, I did not understand certain concepts. The way the agentless monitoring and agent monitoring get used is to be decided by the user, but there is no clarification on the setup phase or on how and why to use the aforementioned options. When it comes to complex environments, the best thing you would like is that the setup should be fairly easy. The product's setup phase is a bit tricky and needs improvement, especially depending on the server environment since the product has a specific file format and a specific configuration form.

One of the things that we would like to improve a little more is the dashboards. The dashboards are good, but I believe that they could improve even more. 

It would be helpful to have a better way of seeing the status of our applications and we can customize the information that is displayed a little bit more.

I would like to see a better way to automate actions based on the events generated by the tool.

I'm not sure I know how I would improve the solution, as I'm mostly pre-sales and sales and I don't have any deep insight into the technical side that could be improved.

There needs to be the capability to modify the code for mobile.

There may be some older technologies that are still covered, however, it's not as plug-and-play. You can still do it, however, it takes slightly more effort for engineers to cover old technologies such as C++, Delphi, and so on. It still can be done. It's just not as easy as, for example, covering Java. Java is just a matter of configuration and clicks.

I would like to see improvements with respect to the visibility of the network. It currently provides visibility on the application side, but not on the network layer.

More visibility and granularity would improve this solution. I would like to see accommodation for millisecond granularity, as we only have one-second granularity now.

In terms of improvements we would like to see, we want to understand what the battery consumption metrics are whenever a user is viewing something in an app. It would be really helpful for us if we could get those details levels.