Keyword Research is Dead. Long Live Keyword Research.

So, the number of (not provided) and (not set) keywords in my Google Analytics reports has now eclipsed the actual keywords reported.


The reason? People being logged into Google (and therefore using https) when they do queries, and therefore not giving away their keyword information.


Does it matter? To me, yes. If only because I like to know them in some special cases, such as if the Bounce Rate is consistently 100% for a keyword, it indicates that the keyword phrase is mis-matched with the content.


Does it matter for keyword research in general? That's a question that it might be a bit early to ask. However, it will skew the results that some suggestion services provide, as there is the possibility that Google users use different keyword types to other people, and are therefore removed from the database.


Arguably, if we assume that logged-in Google users are (a) more internet savvy, and (b) more technical, then if they are removed from the keyword database, a lot of technical traffic goes with them.


Whether that's a good or bad thing, is anyone's guess right now, but I suspect that, in the long term, these things will even out.


I've long suspected that keyword research is a dying beast - to be overtaken by tagging, social media, and various other traffic acquisition strategies that have little to do with search engines in the traditional sense.


However, keywords will remain important for evaluating on-page performance, and for tools such as the AdWords traffic and value estimators built into tools like Market Samurai.


Long live Keyword Research!