How to Measure the Success of Test Automation?

 I have been playing with Test Automation for the beyond 15 years and one thing that continues to come up on numerous occasions is the way to gauge the progress of mechanization.

At first, we used to quantify % of experiments that were robotized. We put forth an objective - X% of experiments that should be robotized inside a time span and afterward we used to approach estimating the advancement, pattern and feel better when we beat the objective.

 learn more - Automation testing

We before long sorted out that this has serious impediments. For instance, we could have 70% experiments that were robotized, yet during a specific relapse cycle, we could execute just 25% - on the grounds that others were not important for that relapse cycle. 

Pretty soon, we wound up with some intriguing (read 'warmed') conversations with partners on how might 70% drop to 25%, etc. mechanization engineers used to gripe that the partners didn't have the foggiest idea what is associated with test computerization and relapse testing. 

Thus, eventually - neither we nor the partners were content with the present circumstance.

It doesn't need to be like this. Allow us to return to the essentials.

learn about- smoke testing

For what reason do we in all actuality do Test Automation?To further develop test proficiency.

OK - thus, what does that mean?It implies that we can save exertion, time and conceivably cost also.

Okay - assuming that is the worth we should get from test robotization - they for what reason would we say we are NOT estimating that?

Rather than estimating % of experiments that were robotized, test computerization measurements ought to zero in on the reserve funds we get from executing those experiments - concerning exertion and time?

It looks basic. In any case, ascertaining the work and time reserve funds through test mechanization execution is everything except straightforward. For showing up at the investment funds -

We really want to realize what amount of time and exertion it requires to execute every manual experiment

We want catch that some place

We want to then plan this to the genuine test robotization scripts that were run and afterward compute this.

This will require some work, joint effort with the manual test group to show up at this.

Also, Read Scriptless test automation

Concurred. However at that point the % experiments that were robotized was not drawing out the worth that test robotization should accomplish, so it's a good idea to transform it to Test Automation Value metric.

We did precisely that and began estimating the test computerization reserve funds. Trust me, it was difficult to gather this data - the test the executives device that we were utilizing didn't promptly uphold catching the data, amassing it and giving us the measurement alongside the pattern. We needed to depend on our past companion MS Excel to show up at this. At times the reserve funds are accomplished north of a few test cycles.

Yet, the outcomes merited the torment. Once, we moved the measurement to computerization reserve funds, we moved the conduct in the right direction.Since we work out the work and time investment funds on what we execute, test mechanization designs vigorously centered around robotizing the experiments -

That were exertion serious, which gave the best reserve funds

That were business basic, since high need experiments get executed all the more frequently during relapse cycle

Experiments that must be executed across various arrangements - OS, Browser, DB, devicecombinations - since they brought about gigantic reserve funds.

Learn more - regression testing tool

Individuals will generally adjust their ways of behaving in view of what gets estimated. At the point when we measure the right things, we empower the right ways of behaving. Kindly offer what are the measurements that checked out and how it changed the conduct in the correct bearing. Much appreciated