It's used in fracking fluids - and also in a LOT of other places, like paints, sealants, cleaning products, et bloody cetera. The shocker would be if they didn't find the stuff.
It's used in many Simple Green products, a LOT of Rustoleum paints, and a lot of others. Minwax, Goo-Gone, Zep, Windex... the list is pretty long. And all it would take would be a home mechanic spilling a bottle of one of those products to get to that same parts-per-trillion levels in their own well water.
The paper suggests that the chemical may have come from a surface-level leak at a nearby well - and that they can't actually tie the chemical to the actual fracking chemicals used at that well.
You raise some good points, but also some further questions that would help validate or invalidate this study. Since this chemical is apparently ubiquitous, is it also found at similar levels in wells not located near fracking sites? Does the concentration vary with distance from fracking sites? Since fracking is relatively recent, is there any historical data on chemical concentrations in wells?
Note that (from TFA) "an unresolved complex mixture of organic compounds was identified" which included 2-Butoxyethanol, and that "Similar signatures were also observed in flowback from Marcellus Shale gas wells" which seems like a bit of a smoking gun. It is also implied that the reason they can't actually tie the chemical mixture to the specific well is that the operator is not making samples available.
Also from TFA is that the wells that were tested were chosen because they were 'foaming' from the chemicals. I doubt that the well foamed from the date of construction.