Page 3 of 3 FirstFirst 123
Results 31 to 42 of 42

Thread: Local Rank stuff...

  1. #31
    Administrator Chris's Avatar
    Join Date
    Feb 2003
    Location
    East Lansing, MI USA
    Posts
    7,048
    There's something new on google.
    Thats true. But you're posting very specific claims with no proof of any sort. It has only been a couple months, not long enough to do any real experimentation.

    How do you explain, that sites with high PR fall off the google results?
    PR is completely independent of context. Yahoo has a very high PR but when I search for "search engine friendly URLs" Yahoo is not the #1 site. With the localrank issue sites that got their high PR from completely unrelated pages may be dropped. Or, perhaps Google simply has a new kind of spam filter. There is no way to know for sure yet, not enough time has passed.

    The patent is also old, and I don't think google's developers needed 3 years to put it into code.
    Like I said. #1 rankings on very competitive phrases from nothing but same-IP incoming links from unrelated sites.

    On the changing the IP stuff.
    You have nothing to back that idea up. Also who cares about money? IP addresses are just about free and if you're seriously making money its worth it to spend $2 for an IP address or $5-$10 for a new hosting account if there is a chance it will make a difference.

    I can understand not wanting to make the expenditure if your site is a poor performer, but then if you don't have successful sites I would question your experience in this field.

    As for your "usage-rank" idea. That just doesn't work. A couple companies have tried that in the past and it doesn't work. It also doesn't explain any recent ranking changes. It doesn't work because some sites are immensely more popular than others. For instance if you searched for something medieval related you might get an Everquest site because that game is so popular, as opposed to a history department page from a university. It just doesn't work to look at raw usage data.

    Now you could like a click-through rates on search results, but then Google has been looking at those (academically they say) for years.
    Chris Beasley - My Guide to Building a Successful Website[size=1]
    Content Sites: ABCDFGHIJKLMNOP|Forums: ABCD EF|Ecommerce: Swords Knives

  2. #32
    Senior Member chromate's Avatar
    Join Date
    Aug 2003
    Location
    UK
    Posts
    2,348
    Well... That's it... Overnight, I've been dropped completely for "carbohydrate counter".

    There's no way to explain this. I just don't get it. What the hell is wrong.



    EDIT: actually, no I haven't. the tool at digitalpoint just didn't pick it up for some reason.
    Last edited by chromate; 02-07-2004 at 04:30 AM.

  3. #33
    Registered
    Join Date
    Jan 2004
    Posts
    183

    hm

    chromate,
    maybe we have to wait for the next index update, which I really hope will be soon(i am also waiting to see if my site would benefit from the link exchanges i did).

    btw. your main page seems too-underoptimized. You have one title, one anchor text pointing to the page itself and that's all. And at the bottom you have merged the two words into: caloriecounter.org instead of calorie-counter.org, and the image has no alt tags.

    You are playing it too safe

  4. #34
    Registered
    Join Date
    Jan 2004
    Posts
    183

    forgot sth

    I saw you have links from 2 pages, placed before yours: one was a post on this forum, and the other was a dmoz page. They should help you, I just don't understand why a page from this forum is ranked before your page.
    Maybe you have underoptimized or sth.

    cheers

  5. #35
    Senior Member chromate's Avatar
    Join Date
    Aug 2003
    Location
    UK
    Posts
    2,348
    Yeah, it doesn't make sense at all. My page was well optimized (see carbohydrate-counter.org/index_keep.php) But after google dumped me to rank #175 I was looking for reasons. I thought it could be over-optimized so I reduced the keyword density big time. After looking at the top 10 sites (that are all really poorly optimized), I decided that it wouldn't make any difference anyway. The following night my ranking increased to #101.

    I'll revert it back to the old index page and see if that makes any difference.

    Looking at the top sites, I should be number #1. Being ranked #101 makes no sense.

  6. #36
    Registered
    Join Date
    Jan 2004
    Posts
    183

    your page

    chromate,
    i don't think your page is optimized for "carbohydrate counter".
    Of course, I am no-one to give SEO advice, but I'll express my opinion and suggestions.

    My suggestions are based on the original google paper + some speculations.

    1. I think that in multiple word searches, google finds for each word the closest other query words and determines the type-of-proximity. I don't think a word participates in MORE THAN ONE multiple word search match. So, a text like this: "carbohydrate counter determines if the carbohydrate content of foods..." the second carbohydrate won't help. Google will determine that the first carbohydrate is closer to the counter and only use this pair to add to the score(the "counter" won't be used in another match to boost the score). Moreover, if the document contains only 5 "counter" words, and let's say 20 "carbohydrate" words, this might indicate that the document is not about the search phrase, although this is pure speculation on my side. I suggest for multiple word queries like "carbohydrate counter" to have the same number (or close numbers) of "carbohydrate" and "counter" words in the text(best is to be phrase matches). If you have to use carbohydrate without counter, you can use counter somewhere next to this to actually indicate "carbohydrate counter" (for example: the carbohydrate content of the foods is determined by our counter..."). This will make the text more readable, because it will eliminate all unnecessary "carbohydrate" usage.

    2. Google uses one bit for capitalization. Wherever possible capitalize and use "Carbohydrate Counter", not "carbohydrate counter".

    3. I don't think keyword density plays a role in the ranking. Judging from the original paper, it is the number of times a word is used in the document for every TYPE OF HIT. Here's what I mean:
    Google's indexer saves for each word hit: capitalization bit, "relative font size", word position. The relative font size is very interesting. The first google used 3 bits for font size(8 possible values from 0-7). 7 meant a fancy hit (title, url, anchor text..). The other values indicate the relative size of the used font. For example, the text with the biggest font is ranked as 7. When ranking a document, google gets the number of hits FOR EACH HIT TYPE. For example:
    fancy hits: 8 times
    fancy subtypes..
    biggest font: 2 times
    ...
    smallest font: 5 times.

    Next google determines count-weight for each type of hit (I'll cite the paper):
    "Google counts the number of hits of each type in the hit list. Then every count is converted into a count-weight. Count-weights increase LINEARLY WITH COUNTS AT FIRST BUT QUICKLY TAPER OFF so that more than a certain count will not help. We take the dot product of the vector of count-weights with the vector of type-weights to compute an IR score for the document. Finally, the IR score is combined with PageRank to give a final rank to the document."
    How do you interpret this? I interpret it like that: For every type of hit, initially the count weight increase linearly, after that it increases by a decreasing step, and eventually does not increase. Let's say that for the biggest font type of hit, the first 3 hits are counted by 1, hits from the 4th to the 8th increase the ranking a little(for example the 4th may be counted 0.5, the 5th 0.25,the 6the 0.12..), and after that you can repeat the keyword(s) as many time as you like but it won't improve the score

    What do most people do? They stuff one type of hit list and neglect all the others. But every type of hit exhausts its influence at some time. I suggest spreading the keywords to different types of hits. Let's say like this:
    count hit type
    2-5 outgoing links
    2-5 image alt tags
    1-3 biggest font
    2-10 next to the biggest font
    ..
    You get the picture? I suggest that stuffing the text with "SAME FONT" keywords won't help too much.

    I think that bold/italic fonts might be some subtypes of the fonts, for example: big font bold, big font, smaller font bold, smaller font, smallest font bold, smallest font or something. By spreading the keywords to different types of hits (fonts, fancy hits) the page will look much more user friendly and search engine ranking friendly.

    In your case, why don't you bold the text at the bottom, and instead of caloriecounter.org use bolded "Calorie-Counter.org". Put alt to the first image. Use "carbohydrate counter" in links. Write more content. It will be links using "carbohydrate counter" to your content. Here's a couple of suggestions for content:

    1) Something with Calorie Counter and yogurt. The carbs in plain yogurt are 3 times less than USDA says. It's because the lactofermentation probiotics eat the lactose(type of milk sugar) and turn it into lactic acid. USDA determines carbs "BY DIFFERENCE", which means it counts everything else, and ASSUMES that the rest is carbs. In the case of plain yogurt, 2/3 of the carbs are already eaten and 2/3 of the resulting "carbs by difference" is lactic acid There was one research study that determined that from 2 groups put on isocaloric diets(the same number of calories) the group that consumed yogurt lost more weight (they speculate it's because of the Calcium content);] You can even manually divide the carb content of all plain yogurts in your DB by 3.
    2) you can write an article on "your carbohydrate counter showing net carbs" or carbs without the fiber. Fiber is labeled a carb, but it is not digestible. My program has presubtracted all fiber from the total carbs(by difference). I even found 3 buggy foods, yeilding negative carb grams :] (SR16)
    3) I can think of some more content for you if you give me some incentive, like a link or sth :] I have been low-carbing for 3 years, and have consulted more than 3000 people and know everything about low-carb diets.


    Now let's repeat:
    1) use capitalization for the keywords
    2) use bold wherever possible(bold vs not bold = different type of hits, more room for ranking formula exhaustment). So some bolded, and some not bolded hits.
    3) erase all other "carbohydrate" hits, use only "Carbohydrate Counter".
    4) put alt tags. you can put some image at the top, and use its alt tags for ranking.
    5) fix that font and carbohydratecounter.org at the bottom to <b>Carbohydrate-Counter.org</b>
    6) add more images with alt tags, biggest font hits
    7) add content using "Carbohydrate Counter" in the anchor text
    8) search for "original google paper" and read it, it is simple and very very interesting

    Hope that helps.
    Cheers, I'll have to immerse in a Prilepin's table powerlifting article :]

  7. #37
    Senior Member chromate's Avatar
    Join Date
    Aug 2003
    Location
    UK
    Posts
    2,348
    Hmmm.. Interesting. You need to remember that it's all relative to the competition though. With my current optimization, I should still be at least on the first page and at the VERY least, better than on the 10th page So whilst you make some interesting points, it's not really enough to solve the main problem of why Google have ranked me on the 10th page

    I've been put their for a reason, and I don't think lack of optimization is a contributory factor really. Since November until this last update, I've been ranked #3 / #2. That made sense. Something has come into play in this latest update that's just not right. Problem is, I don't know what and no one else seems to either!

  8. #38
    Registered Member incka's Avatar
    Join Date
    Aug 2003
    Location
    Wakefield, UK, EU
    Posts
    3,801
    You know Chris charges $1000 for an SEO report like that...

  9. #39
    Registered Member incka's Avatar
    Join Date
    Aug 2003
    Location
    Wakefield, UK, EU
    Posts
    3,801
    [DUPLICATE POST - PLEASE REMOVE]

  10. #40
    Administrator Chris's Avatar
    Join Date
    Feb 2003
    Location
    East Lansing, MI USA
    Posts
    7,048
    My reports are 10+ pages.
    Chris Beasley - My Guide to Building a Successful Website[size=1]
    Content Sites: ABCDFGHIJKLMNOP|Forums: ABCD EF|Ecommerce: Swords Knives

  11. #41
    Registered
    Join Date
    Jan 2004
    Posts
    183

    :]

    Size DOES matter

  12. #42
    Senior Member chromate's Avatar
    Join Date
    Aug 2003
    Location
    UK
    Posts
    2,348
    Especially when you're paying $1k lol

Similar Threads

  1. how soon can a new site get a page rank?
    By jimca in forum General Promotion
    Replies: 14
    Last Post: 11-24-2004, 06:16 PM
  2. how is this possible on google
    By Kyle in forum Search Engine Optimization
    Replies: 9
    Last Post: 01-28-2004, 05:40 AM
  3. Going local
    By Percept in forum General Chat
    Replies: 7
    Last Post: 01-03-2004, 07:30 AM
  4. Media Habit - page rank 7
    By Cloughie in forum Search Engine Optimization
    Replies: 4
    Last Post: 12-13-2003, 01:01 AM
  5. increasing page rank
    By Cloughie in forum Search Engine Optimization
    Replies: 2
    Last Post: 10-02-2003, 08:11 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •