1. Advertising
    y u no do it?

    Advertising (learn more)

    Advertise virtually anything here, with CPM banner ads, CPM email ads and CPC contextual links. You can target relevant areas of the site and show ads based on geographical location of the user if you wish.

    Starts at just $1 per CPM or $0.10 per CPC.

McDar Experiment

Discussion in 'General Marketing' started by compar, Apr 5, 2004.

Thread Status:
Not open for further replies.
  1. irish_ladie

    irish_ladie Guest

    Messages:
    17
    Likes Received:
    0
    Best Answers:
    0
    Trophy Points:
    0
    #21
    Hi McDar - great tool - I love it!
     
    irish_ladie, Apr 5, 2004 IP
  2. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #22
    Bob,

    (?) What are you asking me to do? This is like asking me to list how I think Google should order their results.

    I do not now nor have I ever claimed to understand HOW or WHY Google decides to order their results.

    I made a tool that gives allinanchor results as Google reports them.

    Googles own definition for allinanchor -

    "Allinanchor - Google Advanced Operator. Here are a few commands you can use to help examine the web pages of your site.

    allinanchor: your keywords

    Optimized anchor text is one of the most powerful tools you can use to rank high on Google.

    Use the allinanchor command to find out how you rank.

    This command will show you how rank based on the Anchor Text of your web page."

    actually,
    Now that I re-read that last line of that quote, it makes no sense...

    but that was how Google had it in the Advanced Operators pages.

    Hi Irish Ladie,

    It's nice to run into folks I know. I wondered where everybody went!

    Caryl
     
    mcdar, Apr 5, 2004 IP
  3. Mr T

    Mr T Guest

    Messages:
    62
    Likes Received:
    6
    Best Answers:
    0
    Trophy Points:
    0
    #23
    How about this one?

    'Gazumping' for my house buying site. allinanchor list is pretty much top ten on google search.
     

    Attached Files:

    Mr T, Apr 5, 2004 IP
  4. sarahk

    sarahk iTamer Staff

    Messages:
    28,494
    Likes Received:
    4,457
    Best Answers:
    123
    Trophy Points:
    665
    #24
    Hi Caryl

    A cool tool, thanks.
    I just did a check and one of the sites that poped up was:
    9. home.earthlink.net 240000 11 21

    Interesting numbers because it didn't allow for the subdomain ie home.earthlink.net/~sub/

    I don't know if you can accommodate that but it'd be a "nice to have"

    Sarah
     
    sarahk, Apr 5, 2004 IP
  5. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #25
    I think you missunderstand what I've asked. I have simply said the search I did looks wierd. The result of the allinanchor don't look ordered. You were/are of the opinion that they are odrered. So I said give me a search term that you used to get a result that convinced you, or lead you to believe, that the results are ranked according to highest number or percentage of anchor text?
     
    compar, Apr 5, 2004 IP
  6. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #26
    Hi Bob,

    I still don't think I understand what you are asking.

    I have said that I see a high correlation, in highly targeted keywords, between the actual results and the allinanchor results. (Mr. T's list is the closest I've seen)

    I have seen that with less competitive keywords, many of the top ten listings are not even in the top 100 allinanchor.

    First example: Sleeping Bags

    Second example: Queen Size Sleeping Bags

    ALSO NOTE: I added an extra column with the PR for each site to the first example, just for the sake of discussion. (I looked them up by hand)

    Compare mcdar.com numbers with those of number 5. www.mountain-equipment.co.uk. Allinanchor is where mcdar.com falls short in this example. All I can say is that since I have made a concerted effort to get backlinks with "sleeping bags" in the anchor text, we have finally started to move up.

    Caryl
     
    mcdar, Apr 5, 2004 IP
  7. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #27
    I think you are tending to see what you want the results to indicate. In the Sleeping Bags search look at the site in third place. www.cyber-north.com

    It has 15 backlinks in total -- I know that is not necessarly the total number of links -- but you have 124 backlinks. So it is hard for me to conceive that cyber-north.com has so many more links using 'sleeping bags' for anchor text that it ranks 3 and you rank 228.

    Logic would indicate that you have 8 times more links than they do. So even if cyber-north had 100% of their links using 'sleeping bags' you would only have to have that term on 15% of you backlinks to beat them. How can they be in third and you in 228 under these circumstances? There has to be something else going on here.
     
    compar, Apr 6, 2004 IP
  8. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #28
    Now this result has me really confused. The top line says your site has 124 backlinks, but then when you are listed at #2 in the results it reports your backlinks at 3.

    What is all this about? Have I been missing something? Is there a difference between the backlinks reported on the top line for the site you used in the search dialogue and the backlinks reported for sites in the SERPs??????
     
    compar, Apr 6, 2004 IP
  9. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #29
    Upon further investigation I see the problem. You are only reporting the root domain of each page in the SERPs for any given search term or keyword. The strange looking numbers don't belong to the domain or home page as one would assume. They belong to some inner page.

    I think the results would become more meaningful if you reported the full URL of the pages in the SERPs.
     
    compar, Apr 6, 2004 IP
  10. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #30
    Hi Bob,

    The second column reports the number of links pointing to the page returned in the results. This is not neccessarily the main index page. A mouse over the main website name will reveal which page came up in the actual results.

    So, our main index page has 124 backlinks. The page that came up in #2 has 3 reported backlinks pointing to it.

    Bob, please understand that in making this tool, I in no way intended to say that placement in the serps was totally dependent on your allinanchor placement. Both you and I know that Googles algo is MUCH more complicated than that!

    AGAIN, the tool simply returns information (Googles own results) in a table so those results can be compared side by side. IT ONLY REPORTS GOOGLE'S RESULTS.

    I don't know what the results mean anymore than you do. I am also trying to make sense of the data. Anyone who is at all familiar with statistics, understands that if there is a high correlation between two results, it is worth looking into. BUT, correlation DOES not prove cause and effect.

    There is a high correlation between the moon being out and darkness. Yet, it would be wrong to conclude that the moon causes darkness, wouldn't it.

    This is just a tool that presents Google's results in such a way that it MAY hold some kind of clue. Then again, it may not. I really don't know.
     
    mcdar, Apr 6, 2004 IP
  11. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #31
    Bob,

    I am sorry that I have not made it clear on the tool. In the first column of the results, it shows the main website url. This was done because many of the pages returned were too long to actually fit in the field.

    So, what shows in the field is just the main site url. A mouse over the url will show you the actual page returned.

    As indicated in the ledgend,
    - The "Pages" column is the total pages for the entire site.
    - The backlinks reported are number of backlinks to the page returned in the results.
    - The allinanchor is the first occurance of the site in the top 100 allinanchor results.
     
    mcdar, Apr 6, 2004 IP
  12. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #32
    Some how it seems that you think I'm attacking the tools. I'm not. I'm only trying to understand the results and what they may indicate.

    And BTW you have it wrong. The moon doesn't cause darkness. It is darkness that causes the moon. When it gets dark the moon becomes visible. Cause and effect.

    Let me say this first. I have checked you results manually upon a number of occasions and they are always correct or accurate. Now the the question is what can we learn from them or what do they indicate?

    Here are some of things I think I have learned from them.

    1. I think a lot of people, including myself, have always concentrated on and assumed that the home page was the most important page to optimize. From your results it would appear that Google frequently returns interior pages and these pages frequently have very few links pointing directly at them.

    Possible conclusion: If you had a few anchor text specific links pointed at your internal pages you could probably rank very well. Even with sleeping bags, which I guess is a medium popular search term, the pages that show in the SERP do not have many backlinks.

    2. To search on one's home page may be misleading. For instance you have 124 backlinks to your home page but only 3 pointing to the page the shows up for "queen size sleeping bags". So what is the correlation between that and the actual links that the pages in the SERP have. I would guess none.

    Possible conclusion: A more meaning full results page will involve searching on the actual URL of the page that you want displayed for a particular search term.

    I've run out of other ideas and possible conclusions. I'd be interested in what others think.
     
    compar, Apr 6, 2004 IP
  13. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #33
    Hi Bob,

    The search on the "Queen Size Sleeping Bags" was an attempt to show that less competitve searches do not neccessarily show allinanchor playing a big role (or displaying any high correlation).

    However, in looking over the results it is interesting to evaluate the top 3 sites.

    Example

    If I add PR values plus the total reported backlinks from HotBot, Google, Yahoo, Alltheweb, AltaVista and MSN. It does tell more of a story here (maybe)
     
    mcdar, Apr 7, 2004 IP
  14. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #34
    I think you may have hit on something here Bob.

    It may be far easier to optimize a new page to place high in the serps than to "change" direction of a current page.

    My point being that, the main page of our site has PR6 and 124 Backlinks. HOWEVER, out of the 124 backlinks, a very low percentage of the links contain "Sleeping Bags" in the anchor text. To turn that "theme" around would take a great many new links with the keywords in the anchor.

    I have always "assumed" that your main page carried the greatest weight and there fore the the greater chance of breaking into the top ten. BUT maybe not!

    If I created a new page and linked to it directly from the main page, it would soon have a PR5 (not bad). If I started out from the beginning with ALL links to this new page containing the anchor text then allinanchor should rank quite high as well. Then all I would need to do is be sure to acquire several backlinks to that new page with > PR4. This should get me much closer to the top.

    So, I am off to experiment! ( I will continue my quest with the main page but focus on the new page as well).

    I will report in with results.

    Caryl
     
    mcdar, Apr 7, 2004 IP
  15. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #35
    Caryl,

    I'm not sure the page has to be new. Why not take the existing sleeping bags page and just increase the number of links to it with the appropriate anchor text.

    Some people have talked for years about optimizing every page in your site and I have always assumed that this was not necessary. But your results tend to show that for all but the most competitive areas you should indeed be optimizing the specific page with the item content.

    If you do some checking for very competitive terms like 'buy viagra' for instance, you will find that all the pages in the top ten are index pages. Because in this case the item is competitive enough that people are dedicating entire sites to it.

    But it is doubtful that anyone has dedicated a site to sleeping bags, so in this case you should optimize the sleeping bag page within your site, and by the looks of things you wouldn't need many backlinks to Google bomb that page to the top.
     
    compar, Apr 7, 2004 IP
  16. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #36
    Right on Bob!

    However, those pages are actually for a single item! I do already have a page that does well for one brand of Sleeping Bags, but if someone is looking in general, I would like them to land on a page that shows more than one brand.

    I just picked up a new line so I am going to make a page that includes everything. It really shouldn't take long to establish the new page within the exsisting site.

    I'll be sure to give a blow by blow as to the progress.

    Caryl
     
    mcdar, Apr 7, 2004 IP
  17. mcdar

    mcdar Peon

    Messages:
    1,831
    Likes Received:
    110
    Best Answers:
    0
    Trophy Points:
    0
    #37
    OK - New page up!

    It has two internal links pointing to the page. One PR6 and one PR5
    It has 5 external links pointing to it. All PR5s.

    Now, wait for spidering and indexing.
     
    mcdar, Apr 7, 2004 IP
  18. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #38
    Caryl,

    What is the URL? I'll put up 45 links all PR5 or PR6 to it and we will see if we can drive it to the top of the SERP. I assume the anchor text should be 'sleeping bags'. If we drive it into first place we will talk about how you can return the favour.
     
    compar, Apr 7, 2004 IP
  19. digitalpoint

    digitalpoint Overlord of no one Staff

    Messages:
    38,333
    Likes Received:
    2,613
    Best Answers:
    462
    Trophy Points:
    710
    Digital Goods:
    29
    #39
    We giving away free links now? {I'll wait in line} haha j/k

    - Shawn
     
    digitalpoint, Apr 7, 2004 IP
  20. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #40
    I already gave you a free link. Not 45 mind you, but still a good one.

    I think this is an interesting experiment. Your Keyword Suggestion tool reports just over 700 searches per pay on 'sleeping bags'. That's not [b[high[/b] volume but it is still substantial. It will be interesting to see if we can Google bomb this page to the top with approximately 50 links with the same anchor text. If we can it will answer a lot of questions that are being discussed in various threads on this forum right now.

    If we can't then it is back to the drawing board for many of us.
     
    compar, Apr 7, 2004 IP
Thread Status:
Not open for further replies.