I think the chart supports what people like Ron Wolf have been saying for decades. Higher picks get used at the high value positions - QB, LT, DE/DL and CB. THose positions seem to have the highest percentages of higher draft choices in the 2-deep.
So, use some (say 10-15) of your pre-draft visits on those higher value guys that you expect might figure in to the first round or two. After that, use some on injured or questionable character guys that you want to get a read on (say another 5-8). Use most of the rest on players at positions that are more likely to contribute from the UDFA ranks - OG, OC, ILB, S, FB, LS, P, K.
Of course there is nothing exact about any of this but since the numbers do show that overall less than about 20-25% of your 2-deep comes from the UDFA ranks, it seems wrong to use 80% of your visits on those guys. The reality is that you will draft roughly 8 guys per team and bring in 10-15 UDFAs after the draft. It's reasonable to conclude that 4-5 of the draft choices will be in your 2-deep within the first few years after they are drafted. Maybe 1-2 of the UDFAs. The draft picks still constitute the significant majority of your contributors for most teams. Seems wise to focus on those guys more than UDFAs.