Even in today´s DevOps-adept world, there is a degree of uncertainty regarding the value of test automation. While some see immediate and obvious, advantages, particularly they come to resolving highly granular, or recurrent, process validations; others perceive setting-up automated scripts to be nearly as painful as legacy iterations.
Consequently, we thought we´d offer some useful bullet points in an attempt to clearly articulate the value proposition. However, before we get started, it´d be most useful to understand just what test automation is, and how its significance applies to at-large development.
To wit; ´test automation´ utilizes external scripts to execute targeted code investigations. The development approach is most usually applied in regression, or deeply-iterative test regimes, but can also be applied to any software sub-system that requires the identification of active/passive errors within a particular codebase, or a chain of integrated processes within an omnibus software complex.
For the layman, while the description would seem to be direct enough, it also suggests a couple of other fuzzy terms requiring further discussion. For example, as a test regime, the expression ´regression´ typically investigates the integrated stability of previous and current software versions, after a codebase has been updated or altered significantly.
This verification value may apply at any level within a system including enhancements, patches, and/or configuration changes. At an ´integrated process´ level, automated testing can also apply, where module-to-module code compliance, stability, and/or performance elements are critical. Either way, however, a general ability to create automated ´fire and forget´ test scripts, can save enormous amounts of time and effort, leading to reduced cost, and enhanced revenue over time.
As a practical matter, however, there are some major areas that automated tests that should exist as first principle regimes including:
back to top