So GDN CPMs will likely increase a material amount depending on number of auction participants (IS bid lost will probably correlate with how much more you'll pa

So GDN CPMs will likely increase a material amount depending on number of auction participants (IS bid lost will probably correlate with how much more you'll pay?) and this kind of gets swept under the rug. Now there isn't any safety net from overbidding, so it becomes way more important to start bids low and work up, then periodically downbid and start the process again in the event other participants have entered/exited the auction. How long until this gets implemented on search ads? <2 years? How will it impact the bid floors & ad rank there? (Kind of rhetorical, since the answer is it will most likely drive up CPC bids and for sure increase effective CPMs. I doubt Hal Varian browses HN.) Doing away with AP and IS signaled that this was the end goal. When this happens, you'll have to rely on more frequent bid adjustments combined with both sampling the paid SERPs and the Top IS metric via Ads API if you want to squeeze out more ROI (here is a business idea for someone if Wordstream/Kensho aren't already developing something). It is great looking at the search queries report and ~40% of your spend doesn't even have an associated query. (not provided) for organic in GA is understandable, but doing similar on paid, come on.