A/A testing is done to determine whether the tool being used to run the experiment is statistically fair. The tool should be reporting that there are no differences in the control or the variation. This test is often used to determine the number of conversions on any page which allows the determination of the base conversion rate. The effectiveness and accuracy of A/B testing software is relative to the level of cost savings a business can see.
The other use of A/A testing is to eliminate any possibility that an outside influence is polluting the test results. A/A tests should be used in this purpose only as the last resort, to avoid unnecessary time loss.
When using A/A testing it is important to see if the software reports any type of statistical significance between the variations. If the software reports any type of statistical significance then it could mean that the software on a mobile platform or app is not properly installed.