|brotherhood of LAN|
| 5:53 pm on Mar 15, 2003 (gmt 0)|
>thank good I didn't do as BOL suggested
why not? Brett did [searchengineworld.com] ;)
| 8:35 pm on Mar 15, 2003 (gmt 0)|
I think it is worth discussing any and all factors, and their importance currently. What has emerged in this thread that Google's algo has vastly changed over the last 12-16 months and PR 3&4 easily beat PR 6&7. While all the factors that go into the algo remain, their weightage has shifted. We need to determine which factors have a greater relevancy and which a lower one.
As far as page size goes, I am with RF1...looks like it's the "Earth is Flat" type: that's the way it looks, so must be!
Damn, I wish Google would update, so I could see the changes I made in inward links, outward links and keyword density.
| 10:49 am on Mar 16, 2003 (gmt 0)|
The Fresh index has just been posted on Google.
Homepage: The one change I made, using "Widgets" only in my anchor text from internal pages has moved me up from no. 180 to no. 78! So, now we know the importance of this. And this is for one of the most competitive keywords: 13,700,000 matches.
PR6 page where I increased the keyword density: still at no. 6; but maybe it's a little early as Google is still showing the old text as part of the description (though it's showing the new title)
|brotherhood of LAN|
| 4:28 pm on Mar 16, 2003 (gmt 0)|
maybe worth checking out how this changes your referrals. You may be up on the single keyword, but how are you doing on the anchor text you removed? Perhaps you have enough outbound links saying the same thing? Will that keep you up there?
I started on a three word phrase, ended up at #1, over time, the two word phrase moved up to the top 5, and nowadays the single keyword (noun) of the phrase is in the top 10. it started at 300.
i guess it pays off to start broad and home in on the final choices of keyword you think work best. only you will know with the keywords you have :)
i know its mostly the anchor text from high PR sites and links from on-topic sites. other than that im not much of an optimiser......the content and links done it for me.
if only it was as easy to change the anchor text of those linking to you eh? :)
| 8:22 pm on Mar 16, 2003 (gmt 0)|
Thanks BOL...I'm still where I was for the 2 & 3 words (top 5), as those links are intact coming from external sites. I added the single word link from internal pages...being in the 70s is not much, but now we know the importance of anchor text as compared to earlier (I was dropping in rank even though my PR was going up). I will be trying outbound links and keyword density optimisation and will report back how it goes.
| 8:30 pm on Mar 16, 2003 (gmt 0)|
does Google only Freshbot PR6 and above?
| 8:38 pm on Mar 16, 2003 (gmt 0)|
|does Google only Freshbot PR6 and above? |
No - but higher PageRank pages and/or ranked sites get crawled for refreshes more often.
| 8:52 pm on Mar 16, 2003 (gmt 0)|
"For comparison, the net average is around 15 to 22k per page. My spidering of the entire ODP database when it was at 1,350,000 urls averaged 17k."
Note the the most common (as opposed to average) page according to Brett is around 6k according to that chart. With pages <32k being so prevalent, they will naturally dominate the SERPs. Statistically speaking >32k pages are rare.
| 11:03 pm on Mar 16, 2003 (gmt 0)|
a quick question while we are discussing the various factors:
how do keywords in the URL, followed by extensions as opposed to / do?
will no 2 score better for the keyword imported, or is it all the same?
| 6:31 am on Mar 17, 2003 (gmt 0)|
|<sarcasm> But hey - a few minutes of searching always beats years of research, observation & experience - so what do I know. </sarcasm> |
|hey, if you publish your research, sticky me the URL please fathom. Can't do research and keep it all to yourself and not allow someone to credit/discredit it. |
Ditto, me too! I always appreciate when people share good information; I love to see the findings of good research papers.
| 8:02 am on Mar 17, 2003 (gmt 0)|
Just before everyone starts shortening their webpages to smaller webpages, to so-called please Google..think about the following:
Are you more likely to link to a page with a summary of a topic, or with the whole topic..?
Also, think about the study Brett did on file size. Which pages are most likely to be listed in DMOZ? Index pages? And which pages naturally get the most backlinks? Index pages? And which pages most likely rank high? Those with most backlinks? Index pages? And are index pages naturally very large? or more likely one-screen fitting?
Google has stated its file size limit: 101 kb. For the rest - if at all file size plays a role - I would rank it one of the last 100 variables.
| 5:25 pm on Mar 19, 2003 (gmt 0)|
Hey, don't stop, I've learned more from this thread than any other one yet....
Let's hope I'm getting the drift...
Keywords density, formats and placement - not necessarily the most important SERP factor, but often an important but neglected one?
One question, I've read elsewere that a good target is for keyword density is between 10 and 20% . Is that true?
| 7:58 pm on Mar 19, 2003 (gmt 0)|
Just noticed the fresh results posted (March 17). I am happy to report that 2 pages where I increased keyword density are up from no. 12 & no. 6 to no. 1 & no 3 respectively!
We'll now see what giving outbound links does to pages with their next update.
2nd Fresh update in a row Google hasn't updated PR5 and below pages on my site...anyone getting PR5 updates whose homepage is PR7 or above?
| 9:34 pm on Mar 19, 2003 (gmt 0)|
I'd like to mention that traffic is surging for the keyword string that went from no. 6 to no. 3...I never realised that top 3 makes such a differance!
|brotherhood of LAN|
| 9:40 pm on Mar 19, 2003 (gmt 0)|
Nice one Namaste
|Just before everyone starts shortening their webpages to smaller webpages, to so-called please Google..think about the following |
agreed vita, didn't mean to start a frenzy of people chopping pages in half ;) I'm sure people have reasoning in their judgement yet....hopefully it gets weighted too. Anyway, I pointed to that thread to point out the science/critical nature of it, its out in the open for people to talk about and not heresay on a board :)
| 10:52 pm on Mar 19, 2003 (gmt 0)|
|Giving outbound links to other sites with decent PR |
That should be "relevant sites with decent PR".
Also check for negative factors, like broken links or linking to sites with penalties. Those things might hold you back.
I've had a similar situation for some time -- I've had sites above me with less relevant content and lower PR. I've never found an explanation.
My most recent experiment is to restructure my site. Until recently I had four links (including a "home" link) at the top of every page. I suddenly got a feeling this might be a bad idea (I think I may have been spreading the honey too thinly on the toast), so I've cut down the links to just two. I'll have to wait until the next dance to find out if this makes a difference, but for a few minutes the other day my site had leapt up three places to #6.
| 2:12 pm on Mar 29, 2003 (gmt 0)|
updating this thread to let you all know that I some pages of mine gained in rank over the last week due to optimisation of the meta-description tab!
I used to think that this parameter was obselete and did not bother much with it. But then I noticed that Freshbot is using meta-description in results displayed, so I decided to tweak the wording and got a right result: page rank 12 went to 9; another page went from 17 to 8.
I think this parameter will only work for fresh results, but that's 20days of results!
I also want to report that I did not experience any gains from outbound links. But before I write those off, will await the main update.
| 2:21 pm on Mar 29, 2003 (gmt 0)|
"I used to think that this parameter was obselete and did not bother much with it. But then I noticed that Freshbot is using meta-description in results displayed, so I decided to tweak the wording and got a right result: page rank 12 went to 9; another page went from 17 to 8."
Yes, you are right. Meta description is of great importance, and I try to have unique descriptions for each page. Meta keywords aren't of great value for ranking. It's just that you should not have meta keywords, that do not appear in the text.
| 8:13 pm on Mar 29, 2003 (gmt 0)|
|you should not have meta keywords, that do not appear in the text |
is this your experience? Have you seen positive results by syncronising them with words on the page? Have you seen a negative impact if they are not syncronised?
| 9:01 pm on Mar 29, 2003 (gmt 0)|
|Google has stated its file size limit: 101 kb. For the rest - if at all file size plays a role - I would rank it one of the last 100 variables. |
I agree - and conceed that many other variable are more "weight relevant".
Rarely though do you actually discuss all variables together, which we really should.
File size - may not be as big a player as links, anchors, and titles, etc. - but it is a player - nonetheless.
| 12:06 am on Mar 30, 2003 (gmt 0)|
"is this your experience? Have you seen positive results by syncronising them with words on the page? Have you seen a negative impact if they are not syncronised?"
About half a year ago I wondered why only a small fraction of my pages where indexed in Google, so did some research on the internet, and somewhere I read, that you should not use meta keywords, that can not be found on the page. Then I adjusted all the pages and sure enough about a month later, all pages where indexed, and ranking well.
|brotherhood of LAN|
| 12:53 am on Mar 30, 2003 (gmt 0)|
just another argument for filesize "mattering"
If a page has 10 words on it, there's less chance of words being off topic, and on the same topic.
If oyu have a 100 words on a page, talking about a particular subject, you can more or less determine what the subject is with the nouns on the page. Chances are most of the words will be needed to make proper sentences to define the topic.
If something is a 1000 words, what's the chances that some of it is off-topic, maybe "hidden text", or something to that extent. Seems you can have a degree of "reliability" if you chose only to index smaller pages.
A 60K page most likely has bells and whistles that have nothing to do with the content, so it can be harder to determine what the "core" topic is.
Just seems that Brett's filesize chart is not just a coincidence. Then again, it might just be google's way of making best use of its resources. It must take longer to find the "ransom note" on a page that is longer.
just another filesize 2cents ;)
hows the site coming along namaste...you still ranking higher for those changes you made?
| 5:04 am on Mar 30, 2003 (gmt 0)|
If i was G, i'd throw in a lightly weighted random sort_order every month. It would increase fair competition (nice rotation of sites), and throw us all off track completely.
We'd eventually have to stop all this, and focus on developing our sites.
I'd also be rich :)
| 9:33 pm on Mar 30, 2003 (gmt 0)|
Well I've cut my homepage k in half. Lets see what this does for ranking.
BOL, the following changes I made since this thread started have yeilded positive results:
1. Inbound links from internal pages featuring the targetted keyword in isolation. Result: ranking imporoves from 180 to 69 overnight! Now am awaiting the main update whereby external sites linking where the keword has been isolated will start counting.
2. Increasing keyword debsity. Result: ranking boost of several positions.
3. Changing meta-description to be more in line with targetted keywords. Result: gain of several positions.
No position gain:
1. Outbound links: no gain in any position...but perhaps this will happen after the main update. Above results are via Freshbot updates
Still to show results:
1. Page size reduction. While freshbot has now listed the new page, it has no boost in ranking. Perhaps it will happen with the main update
2. Alignment of Meta Keywords tag with targetted keywords found on page...just done, lets see what happens.
I might mention that yesterday, on AV, I am now top 10 for my main keyword, which has 5,800,000 results on AV.
I saw a very interesting thread which makes sense about page size: [webmasterworld.com...]
Keep the tips coming and I'll keep reporting what happens as I implement them.
| 1:26 pm on Mar 31, 2003 (gmt 0)|
just saw the Mar 30 update posted on Google...the page that was made lighter is up a few notches under certain keywords.
| This 85 message thread spans 3 pages: < < 85 ( 1 2  ) |