There must be a way because all ranking tools have to use serps without localization.
So you would think, but it's interesting how many of the tool providers haven't even thought about this issue.
That said, it's also an extremely difficult issue to handle, because there are times when Google's results are hyper-localized, so much so that shifting your location just a few miles can change them. There are other times when such extreme localization doesn't appear to be happening, or it's happening on a different set of queries.
See my fairly long post in this thread, from back in February, 2012, which chronicles various tests I've run....
False High Rankings - just our work location http://www.webmasterworld.com/google/4423084.htm [webmasterworld.com]
Note that the default location you can set on the Google serps page provides apparently dependable control... so you may not even have to use proxies. Chances are the results are localized to city-center of the specified city. I remember that a test was run a year or two ago using searchers all over the US connected by telephone, and the locations which distant searchers set manually produced results that were the same as what local searchers got. (If anyone can find the original discussion of that experiment, please feel free to mention it here).
The manual location setting currently is buried in
"Search tools", up at the top of the Google serps page beneath the search box. The tools menu brings up a line of several options, and to the far right are the location settings.
The closest I've gotten to a neutral result where I live (in northern California) is to specify "United States", and it's possible that
countryname may work elsewhere. Some abbreviations (like "US") also work, but that may depend on the sequence of other settings I've tried and when I've flushed cookies, etc. "Anywhere" is not a "recognized location".
I don't know, btw, how useful rankings for a countryname location actually would be to you. It may be that localized results will give you a better idea of what a typical searcher would see for typical searches in other locations.
Possibly, a sophisticated tool could run multiple searches in major locations and determine averages. Something similar may have been done on some commercial tools that "correct" for personalization, but location is the most distinct of all personalization factors. It's hard to "average" it out.
To add one further thought... I've also determined that searchers in different regions of the country are likely to use different vocabulary in some queries... so that's a whole other set of variables.
