Hey @methode any probability regex filters coming to the GSC Search Analytics report any time quickly?
— Andrew Shotland (@localseoguide) August 15, 2015
Nicely it lastly occurred. Google Search Console’s Search Efficiency report now helps common expressions. So that is nice as a result of now you can do cool searches like:
However whereas we waited six years to have the ability to correlate alan bleiweiss with asshat (it’s not causation of us), our crack TechOps crew determined to determine how to do that ourselves.
We website positioning varieties are by no means glad with the free instruments that Google provides us, as a result of with just a little additional work, there’s often there’s a technique to get one thing way more helpful. After all I’m speaking concerning the Google Search Console API. For those who haven’t found out how one can use it but, I strongly suggest you prioritize making it occur. Listed here are some website positioning tech nerd the explanation why:
The API already let’s you do regex searches! Put the API knowledge into one thing like Google Huge Question and you are able to do advanced SQL queries over massive knowledge units and far sooner than through GSC.
You get far more knowledge from the GSC API than within the GSC interface – ask Noah Learner for his latest LocalU deck to see what somebody artistic can do with that knowledge. Right here’s an instance:
GSC exhibits 1,000 key phrases for this web site for the time period “pci”:
Now right here’s what we get utilizing the API with Google Huge Question:
We discovered 3,281 phrases, greater than 2K greater than we get through the GSC interface. In case you are not utilizing the API you could be lacking 66% of the information. How are you going to make selections based mostly on that?
If in case you have arrange an information warehouse you may entry greater than 16 months value of knowledge.
Google’s help of regex is often fairly restricted. For instance, the GSC regex at present doesn’t help damaging lookaheads, or not less than I couldn’t determine how one can do it. We’ve a shopper that unintentionally ranks for a ton of porn queries which obscures numerous the legit queries. There is no such thing as a technique to filter the GSC knowledge inside the interface to take away these queries, which makes it principally unusable.
Pandas (the Python knowledge evaluation library) already helps you to do fancier stuff natively like knowledge frames (pivot tables) and means that you can examine issues collectively inside knowledge frames.
So you may examine two regex filters towards one another to see what subset is performing higher in GBQ or Pandas.
With GSC knowledge loaded into Google Huge Question, you may make the most of analytic capabilities inside Postgres to check time-frames simply to see the place progress is (or isn’t in your question corpus) extra time; by segmenting the information into buckets you may then suss out particulars which are in any other case misplaced to GSC GUI customers. For instance, utilizing a LAG() + OVER clause you may generate a report that exhibits the % month over month change for clicks and impressions for a gaggle of queries like:
For all you Postgres geeks, the code for the way to do that is offered on the LSG Github
If in case you have simply spent a number of hours making an attempt to determine re2 regex, do your self a favor. Discover somebody who can hook you as much as the GSC API. You’ll be happier as will your website positioning.
