'E2E' is often impossible as in most ecosystems as things are constantly changing and you will dependencies out of your control you cannot simulate. The key i

'E2E' is often impossible as in most ecosystems as things are constantly changing and you will dependencies out of your control you cannot simulate. The key is faking the right dependencies with accurate-enough versions to keep test fidelity and speed, keeping the test svelte and fast enough so it can run before you merge code to eliminate the size of change being tested and thus more easily understand the outputs to determine if they represent a false positive or negative and where any problems may be. This also all requires building the infrastructure to spin up a simulated world quickly enough to simulate how the proposed change effects the simulation and then analyze the results which is also pretty hard and can get expensive. Luckily spending time optimizing often helps test speed and can control cost so there can be a good case to make for it but orgs have to be willing to pour engineering hours into that and engineers need to want to do it vs building new things which is typically more enticing.