Quick answer: the 12-word limit includes stop words. That is, Google will only index the first 12 words of a title, regardless of whether or not they are stop words. This conclusion is derived from the fact that searching for this post by the first 12 words–some of which are stop words–returns at least 1
intitle: match, but searching by the 13th word returns 0
intitle: matches. However, a search for the 13th word without the
intitle: operator returns at least 1 result, which suggests that Google indexes all words of a title but only counts the first 12 as title hits. In simple terms, this can be translated into an SEO best practice: Make sure your keywords appear within the first 12 words of your page titles.
In my previous experiment, I hesitantly concluded that Google only indexes the first 12 words in a title. Shortly thereafter, some smartass San Francisco SEO and internet marketing guy named AJ Kohn rebutted my findings by claiming that he successfully queried the lyrics of a certain song by The Police:
I’m not sure this is accurate. I’ve been able to get 13 ‘words’ in a title by doing a search for: de do do do de da da da lyrics
However, this doesn’t necessarily disprove the 12-word limit theory. What it *does* do is indicate that I need to be a bit more clear with my definition of index. For example, in the following sentence:
Google only indexes the first 12 words in a title.
I’m using indexes to mean: parses the content from a document’s title tag and stores each of the parsed words in an inverted index as “title hits” associated with the given document.
In other words, I’m theorizing that Google’s indexer will only give a certain document “credit” for having a given word in its title if that word appears in the first 12 words of the title tag. I’m also assuming that the
intitle: operator is reliably capable of querying this information. That is, Google hasn’t programmed its search engine to return 0 results as a way of refusing to serve
intitle:queries containing more than 12 words.
This use of the word index has nothing to do with how many words or characters are displayed in Google’s SERP titles. In other words, just because Google displays 13 words in a SERP title doesn’t mean that document is actually getting credit for all 13 words. To understand why, you need to have a deep understanding of search engine architecture–which I’m too lazy to provide (and which I’m not convinced I myself have).
Instead, I’ll tell you that everything you’ve just read is merely unique content to help this page get indexed, because this page is actually another experiment…the hypothesis of which can be derived from the title…and the results of which will be posted later.