The standard way to deal with this is to use color calibration. Unfortunately, color accuracy on consumer devices tends to be all over the charts. There is no way to guarantee that the color will be similar, even on multiple of the same model of device, or even on the same exact device over time.
Thus, the best practice is to setup a color calibrated environment that you can verify the content looks correct in. You can then run your encodings for different profiles and compare them in an environment where you know the differences will show up clearly. The results on end user devices may vary based on how good their screens actually are, but building to a standard high quality, calibrated display should produce the best overall outcome across devices.
If you want to get really advanced, you can also generate or obtain ICC profiles of the average for the device screen you want to display the video on. This profile can then be applied to the video on a calibrated display to alter the colors to be similar to that produced by the device. It still won't take in to account pixel density though. If you want to know exactly what it will look like on a display, there is no substitute for actually sending it to that display since pixel layout and density, as well as the surface type and size, make a gigantic difference in the way the video looks.
Using the device itself is also helpful for ensuring that the video can decode and play smoothly within the constraints of the device.