good candidates for automation

1Tests that should be automated:
2Business critical paths - the features or user flows that if they fail, 
3cause a considerable damage to the business.
4Tests that need to be run against every build/release of the application, 
5such as smoke test, sanity test and regression test.
6Tests that need to run against multiple configurations — 
7different OS & Browser combinations.
8Tests that execute the same workflow but use different data for 
9its inputs for each test run e.g. data-driven.
10Tests that involve inputting large volumes of data, such as filling up 
11very long forms.
12Tests that can be used for performance testing, 
13like stress and load tests.
14Tests that take a long time to perform and may need to be run 
15during breaks or overnight.
16Tests during which images must be captured to prove that the application 
17behaved as expected, or to check that a multitude of web pages 
18looks the same on multiple browsers.
19Generally speaking, the more repetitive the test run, the better 
20it is for automation.
21
22Also some tests are not the only candidates for automation. 
23Tasks such as setting up or creating test data for manual exploratory 
24testing are also great candidates for automation.
25
26Tests that should not be automated:
27
28Tests that you will only run only once. The only exception 
29to this rule is that if you want to execute a test with a very large 
30set of data, even if it’s only once, then it makes sense to automate it.
31User experience tests for usability (tests that require a user to 
32respond as to how easy the app is to use).
33Tests that need to be run ASAP. Usually, a new feature which is 
34developed requires a quick feedback so testing it manually at first
35Tests that require ad hoc/random testing based on domain 
36knowledge/expertise - Exploratory Testing.
37Intermittent tests. Tests without predictable results cause more 
38noise that value. To get the best value out of automation the tests 
39must produce predictable and reliable results in order to produce pass 
40and fail conditions.
41Tests that require visual confirmation, however, we can capture page 
42images during automated testing and then have a manual check of the images.
43Test that cannot be 100% automated should not be automated at all, 
44unless doing so will save a considerable amount of time.