Currently, we must write logic separately for both platforms (Android and iOS), which is time consuming and redundant. The platform should support one logic that works on both platforms; of course, the locators must be identified, but I believe this will eliminate duplicate logic writing.Consider this: if you’re writing 20 test cases with some if else on one platform and then need to write the same thing on another, it’s easy to overlook a step/validation. Hope you will comeuppance with some solution
From our past experience, building an automation for 2 platform using a single generic test provide a flaky test that usually need manual repairing, in our research it came to light that it is faster to record an automation once for each platform rather then use a single test both for and trying to fix every step that fails manually.
There are a lot of other considerations as well, a test and elements are made per application and different platform use different locators strategies, you cannot combine on the same element both Android - ID and IOS predicate locator strategis, it will break the self healing capabilities.