Good idea!To add to your proposal,
1) How long player was member of team which has sold him; example - longer player stays with the team ---> bigger percentage from future transfer of that player
IMO the time a player spends on a youth team should count 1.5-2 times that for older players. (i.e. a player with 2 years on a youth team should count as 3 or 4 years on the regular squad). Also,
any player should have to spend at least one entire season with the club to have any positive effect on this %. Perhaps appearances in (competitive) matches could also figure into this, so a player who has hardly ever played in these matches would have less of an effect than a first-team player, regardless of their time with the club.
2) How many transfers has selling team made in the given period of time; example - lots of transfers in certain period of time ---> smaller percentage from players which will be transfered
Agreed, but players transferred within one's own league AND players from one's nation purchased from foreign leagues should not count against the team's transfer tally (or at least count less) for these purposes. I don't know how feasible that would be, but players' nationality and league should count for something.