|Matt Cutts Video: Ranking Content Without Many Links|
How can content be ranked if there aren't many links to it?
Matt Cutts - Google Webmasters Video
trt 1:30 - June 2, 2014
I thought this was very interesting, less for what Matt said but for he didn't say. It seems that, except for the longest of the long-tail, it is nearly impossible to rank without some links.
That means some kind of promotion is needed and it's not what Google's been attempting to make us believe for a couple years, that great content alone will let you rank for the keywords that matter. A strong domain also helps, but that doesn't do anything for the new sites out there.
There is no need to read between the lines here. He visibly struggled to answer the question.
[edited by: Robert_Charlton at 9:19 pm (utc) on Jun 2, 2014]
[edit reason] added video title and made link clickable [/edit]
I don't think MC is telling it all in that clip. Nowhere in that clip does he mention on page / website metrics - such things as how long a user spends on a page, how many pages per session the user views and what happens when they exit the website. The very absence of any mention at all (absolutely none at all) tells you one thing, MC is not telling the whole truth. On page /site metrics matter a massive amount, they tell G so much that they never, ever could ignore them but MC still prattles on about links as if (exceptional cases excluded) they were the Holy Grail.
Let's take the absurd situation where a new site is published on blue widgets. It a 50 page site and it has only one external link to it. On average a user spends 5 minutes on a page and per session, they view 10 pages. Let's also assume that the average page / site metrics for similar pages is a tenth of that.
Is G going to ignore those stats in favour of another site that has 50 external links but has abymyssal on page / site metrics - that assumes of course that they want to feature sites in their SERPS which satisfy their users needs. Of course not. G knows full well that links are nowadays an almost unquantifiable quality factor. But if a 1000 users per day result in on site / page metrics as mentioned above then they know that the site is satisfying their users needs better than all the competition.
Don't get led down a back alley by MC when even a short analysis of the facts show what is really going on. Links can so easily = spam and G has no idea of the difference between spam links and real links - and they know it.
|Nowhere in that clip does he mention on page / website metrics - such things as how long a user spends on a page, how many pages per session the user views and what happens when they exit the website. |
Because they don't matter in this case. Obviously there are no such page metrics because the page is brand new.
I have top-ranked pages for highly competitive keywords with no links on domains that don't even rank in the top 10 for their own names :) All based only on content, so it's perfectly doable, unless you are competing in an area that Google has "sold out" to preferred websites.
Ironically, I have also tried to outrank my own pages with pages from highly developed domains and was never able to succeed. I guess, if Google liked it so much in the first place - it's keeping it at the top for a while.
|The very absence of any mention at all (absolutely none at all) tells you one thing, MC is not telling the whole truth. On page /site metrics matter a massive amount, they tell G so much that they never, ever could ignore them but MC still prattles on about links as if (exceptional cases excluded) they were the Holy Grail. |
Yup -- I have an EMD that doesn't have a single "follow" link to it other than scrapers as of the last time I checked, but yet somehow the domain ranks #1 for the EMD query and other highly relevant queries -- Why? IMO, because when people type in the EMD search or a relevant search it ranks for, the search, more-often-than-not, ends.
I think user behavior and "the search ending" are bigger factors than MC or SEOs give them credit for personally -- IMO Building a site to "end the search" is what SEO should be about. Nothing else.
I think Matt's answer is trying to answer the question of how a site might rank initially just for onpage factors... and in a sense his answer traces why factors like repetition and keyword density quickly became inadequate as signals for search engines to evaluate anything competitive.
He did avoid commenting on the ingredients of Google's "secret sauce" which guide how Google does evaluate onpage content. I'm sure phrase-based indexing enters into it a lot.
The question reminded me of this thread from July 2012, which discusses some comments Google's JohnMu made about the time and signals a search engine needs to recognize that a site is as "fantastic" as claimed....
Search engines need time & other signals to confirm a site is "fantastic"
Essentially, John is talking about link and traffic signals, and I think he's saying that, lacking those, onpage factors alone can't establish a site's competitive worth. That isn't to say that Google doesn't use onpage factors... I think that it uses them quite a bit... but it also wants backlinks for ranking confirmation.
|On average a user spends 5 minutes on a page and per session, they view 10 pages. Let's also assume that the average page / site metrics for similar pages is a tenth of that. |
You mean the 3 people a day that visit the site because it has next to no traffic? Hardly strong data.