The average calls response time can get inconsistency in results of 36% or even sometimes more when comparing results of 2 runs of the same test. This make it difficult to analyze our web application for any type of performance degradation from one build to another.
As far as we understand, Neoload executes scripts step by step but not waiting the result from the 1st step and going immediately to the second one.
Let’s say we have in one page:
All those steps executed one by one (meaning Step1 executed 1st, Step2 executed 2nd and Step3 executed 3rd but have different response time).
When we run Check User Path, for example, we can get Step3 response first, Step2 response 2nd and Step1 response 3rd.
Considering this can bring to inconsistent results from one run to another, and if so, can we change default behavior of Neoload to execute steps sequentially instead of in parallel?
I'm not sure to understand what your 3 steps are. If your steps are requests in under a page then yes it is played in parallel and step3 could be played before step2.
But it is how a browser works. Requests under a page are usually resources like JS, images etc... so the order to execute them does not matter.
Now if your steps are pages or transactions then they are run sequentially. That means that NeoLoad executes them in the same order than in your script.
If your steps are requests and you would like them replayed in the same order you should copy them individually on separate pages.