How many customers were retained because you decreased response times by 100ms? How many customers were gained because of that slick UI you created?
These two things can be precisely measured, provided you have the traffic. And you should be measuring them anyway, because that slick UI might be hurting things.
A/B test all the things. Then you'll know both your value, and whether any specific improvement you made is worthwhile.
A salaried employee is never going to gain anything from this except a small bonus, a marginally higher salary increase than his or her peers, or a pat on the back.
These two things can be precisely measured, provided you have the traffic. And you should be measuring them anyway, because that slick UI might be hurting things.
A/B test all the things. Then you'll know both your value, and whether any specific improvement you made is worthwhile.