How do you compare the effort of making a recording versus scripting a test from scratch using your Structural Testing commands?

Actually we recommend a "hybrid solution" as follows: Record the test from live and then hand-modify the script as needed.

Putting this in verysimple terms we mean: Record the test with the recorder and let the software do all of the heavy lifting, and then hack on the resulting script until you have it doing what you want.

This is a point of view that places a premium on getting results quickly and efficiently, perhaps with less regard to the purity of the method or the outcome. We have been doing this for quite some time now and the scripts that get done this way may be a little messy and may not be perfectly formatted but the work extremely reliably.

That's the actual goal, isn't it?

We use this exact same method in our "fixed price" script development work, in which for a small fixed fee we work from your use-case or even your outline of the test sequece and produce an AJAX-reliable, durable (non-brittle) test that plays back with high fidelity (excellent timing and performance accuracy).

The eValid Team