That's an excellent yield response. If you could do that year after year it would be a no-brainer. 

I'm assuming by your description that you did a side-by-side comparison. If this is correct, how did you account for the inherent yield variabilty within the field. I ask this because if you had an inherent 6 bu. variance - which would not be uncommon - a 10 bu. increase doesn't look quite so good. On the other hand, if the 10 bu. was over and above the natural variability, the net return makes the decision to apply an easy one.