yes that be interesting to try , if for no other reason than to see if can work , not sure about the extra chip count to achieve ,

in the end i like reliable and consistent in keyboards
if Jerson says he has seen the 1 pin adc failures , then its likely that the approach has been used in commercial products. but with all things its depends on the usage as well , enviroment etc

the application i have is not in wet environment,but is used outside and is covered , key usage for the membrane is for some programming and not heavy usage even then.

the application requires serial lcd display , keypad , serial port to pc via serial to usb adaptor , 3 other special allocated ports for a outputs , 4 other spi devices on a bus, i still will have 3 other switches which will not be on keypad , bottom line is i/o pins are in short supply unless i start to look at a larger pic , which maybe the way to go, but i am resistant to go purchase another pics at this point in the design

.