Also, the sort of formula shown in that article makes my skin crawl. Not because it's hard – it's mostly a set of sums based on conditional lookups with rounding – but because by the time you've written something like that, Excel is almost certainly the wrong paradigm for your data. It's a terrible IDE; making minor changes to that...thing will half as happily pop an error message and junk it as it will do the right thing. Similarly, you can't exactly git excel formulae alone, and the mixing of data and analysis inevitably causes chaos at some point – just look at all of the examples of financial disasters caused by it. I carte-blanch refuse to touch anybody's attempts at doing stats with it (in a medical science context), unless it also comes with R, sas or stata script..

What the article's really trying to say is "Being good at Excel implies you have a logical mind and the ability to see a way through problems. This is something that others lack".