It was a surprising observation that the time at satellites in earth orbit differs much more from the time at earth as expected. How great is the daily difference actual?
There is a small relativistic effect, but I think it is probably automatically account for in any established system. Perhaps you read about an error in someone's system?