| 10:27 am on Aug 5, 2014 (gmt 0)|
Have you looked at the keyword density for the top ranking results in Google? Did you see any correlation? Was that correlation strong enough to suggest possible causation?
| 4:01 pm on Aug 5, 2014 (gmt 0)|
Build your page / site for your desired audience.
Make it so much better than your competition's pages for your target audience that it is a no-brainer that they would want to go to your page instead of your competitors.
Google has gone way beyond simple metrics that can be easily games (such as keyword density) and looks much more at things like user interactions and social approval.
| 5:57 pm on Aug 5, 2014 (gmt 0)|
Many times I've searched in Google lately for a keyword phrase or keyword and done a text search for that phrase or word. The result? Often times there isn't a SINGLE INSTANCE of that keyword on the page. That should tell you everything you need to know about keyword density. Oh, also these are top results pages I was looking at. When I approach an article/page now, my keyword usage is the least of my concerns. Also consider the fact that my success lately has plummeted off the face of the earth. However my content that still ranks is written with the mindset that keywords meaning next to nothing in Google.
| 12:32 am on Aug 6, 2014 (gmt 0)|
Thank you all. Now I know I should not pursuit keyword density when I write content, just focus on audience.
| 2:12 am on Aug 6, 2014 (gmt 0)|
amilylorry, in my experience, it helps to have core targeted vocabulary on the page, to such a degree that I'd say it's essential.
I've seen a few cases where Google ranked a page that apparently didn't contain the targeted vocabulary, but those were special cases, way outside the norm... often blackhat spam. I can imagine with various kinds of query rewriting now happening, or for very special kinds of searches, that this might have changed recently, but I tend to doubt that it's generally true. I haven't seen it myself.
Excessive repetition, though, isn't rewarded, if that's what you're asking.
And exact match on long tail queries isn't what Google is looking for. There Google is looking for concepts... but for the overwhelming number of searches and pages, the presence of core vocabulary IMO is generally still an important part of that.
| 8:51 am on Aug 6, 2014 (gmt 0)|
@ Robert Charlton:
I am glad you brought that up.
One of the googlers (John Mueller?) said of Hummingbird that google is moving "from strings to things."
While I think this has been ongoing for sometime now, my interpretation is that Hummingbird took a new step in going from matching strings of text to matching named entities.
Would you agree that Humingbird took the implementation of named entities and kicked it up a notch (or two)?
| 2:17 pm on Aug 6, 2014 (gmt 0)|
|Excessive repetition, though, isn't rewarded, |
Sorry, I have to completely disagree with you on that one. I have some competitors and all they have done for years is repeat keyword after keyword after keyword in the titlebar, in the description, on-page, ad nauseam and they still rank all over the first page.
Admitedly anyone visiting their pages would probably think "WTF?" but the fact is that G still does not have this under control.
| 2:58 pm on Aug 6, 2014 (gmt 0)|
|While I think this has been ongoing for sometime now, my interpretation is that Hummingbird took a new step in going from matching strings of text to matching named entities. |
For anyone curious about how well this works, there's some pretty good demos:
Academic research: [cogcomp.cs.illinois.edu...]
A commercial product: [alchemyapi.com...]
| 3:34 pm on Aug 6, 2014 (gmt 0)|
Why are you sure that your competitors success is mainly due to KW-density?
I remember JM from Google answering this question clearly with "no". Think about it and read the posts here and you will find many good reasons for it.
| 8:11 am on Aug 13, 2014 (gmt 0)|
Keywords are important on the page, but do not overdo it. They should be in moderation. Each page can have a different number of keywords. Better to write text on the page, and then check the keywords <snip> and only then decide how much they should be used in the text. Check the same pages of competitors, and you'll see what percentage of cells.
[edited by: goodroi at 4:34 pm (utc) on Aug 13, 2014]
[edit reason] Welcome to WebmasterWorld, please re-read the rules [/edit]