I use it Dynamic Creative all the time, but don’t feel pressure to use all 10 / 5 variants unless you think the additional variant is worthwhile. There is a tax to FB learning something doesn’t work well and the benefit of that learning should outweigh the tax. Also, the learnings from variant performance in the breakdowns are certainly interesting, but I haven’t seen any confirmation that the permutations are ‘balanced’ and that a given image, for example, will have the same distribution of Ad Headlines as another image. So it could be that your image is doing great, or it could be that is just got paired with the best Ad Headline more often.