I love the sentence “More evidence should not mean proportionally more text.” So many tools perform statistical SERP analysis without context and give random metrics that are far from real usage, just because they are statistically close.
Thanks for such great work!