Page 3 of 4 FirstFirst 1234 LastLast
Results 31 to 45 of 46

Thread: Compete.com an Alexa competitor

  1. #31
    Site Contributor KLB's Avatar
    Join Date
    Feb 2006
    Location
    Saco Maine
    Posts
    1,181
    Quote Originally Posted by rpanella View Post
    The fact these competitors to Alexa are coming out is great though, because by using them all together (triangulating the data if you will), it is likely possible to get more accurate figures than any one service could provide by itself. Also, just having competition will push them to provide more accurate data.
    1) "Triangulating" using bad data will not result in good data--it just can't happen.

    2) What is important is not an exact traffic level for a specific site, but its traffic levels compared to other sites. For the comparison to be relevant the numbers MUST NOT be skewed do to serious flaws in methodology.

    3) Alexa's methodology is flawed beyond repair. It relies upon scientifically and statistically invalid methods of data capture. Sample data is not collected from random individuals, rather ones who seek out and install Alexa's toolbar. This requires the individual to know about Alexa, be able to use the toolbar and be willing to install it. The only way Alexa's stats could be fixed would be to abandon the toolbar and shift to buying click stream data directly from ISPs (as Compete is doing). Alexa's data and toolbar is nothing more than a marketing scheme used by Amazon.com disguised as traffic data. Quite simply in the case of Alexa, the emperor has no clothes.

    4) GA will be inaccurate (although very useful) because it relies on the browser supporting JavaScript and allowing cookies.

    5) Comparing server log stats between sites will also be inaccurate because of different methodologies to detect and weed out robots/bots/spiders and different calculations of defining what is a "unique".

    Now granted some methods will return better data than others, but one should never be deluded into believing one method is truly accurate.
    Ken Barbalace - EnvironmentalChemistry.com (Environmental Careers, Blog)
    InternetSAR.org: Volunteers Assisting Search and Rescue via the Internet
    My Firefox Theme Classic Compact: Based onFirefox's classic theme but uses much less window space

  2. #32
    Registered
    Join Date
    Aug 2006
    Location
    Sacramento, CA
    Posts
    208
    Quote Originally Posted by KLB View Post
    1) "Triangulating" using bad data will not result in good data--it just can't happen.
    By taking 3 small samples of data, you create a much bigger sample size, which will reduce the skew of each sample individually. This is basic statistics, which you do not seem to understand. Everything you have said in this thread is along the lines of "Compete knows my exact traffic therefore they are completely accurate, possibly even moreso than direct metrics such as google analytics".
    ________
    Amateur Video
    Last edited by rpanella; 03-17-2011 at 10:46 AM.

  3. #33
    Registered
    Join Date
    Aug 2006
    Location
    Sacramento, CA
    Posts
    208
    This thread has gotten very off topic but here is a useful site I found that lets you quickly compare traffic graphs between different sites with all 3 of the services: Alexa, Compete, and Quantcast.

    http://www.attentionmeter.com
    ________
    BABE STRIPPING
    Last edited by rpanella; 03-17-2011 at 10:47 AM.

  4. #34
    Site Contributor KLB's Avatar
    Join Date
    Feb 2006
    Location
    Saco Maine
    Posts
    1,181
    Quote Originally Posted by rpanella View Post
    By taking 3 small samples of data, you create a much bigger sample size, which will reduce the skew of each sample individually. This is basic statistics, which you do not seem to understand.
    But if that data is skewed, it doesn't matter how many data points you pick. the results will still be skewed. There is no way around it.

    Quote Originally Posted by rpanella View Post
    Everything you have said in this thread is along the lines of "Compete knows my exact traffic therefore they are completely accurate, possibly even moreso than direct metrics such as google analytics".
    NO, that is what you may be reading into what I said, but it is not what I'm saying. What I'm saying is that Compete is the most accurate indirect traffic reporting service I have seen. Notice the key word "indirect". GA and server logs are direct traffic reporting tools.

    I'm also trying to get people to realize that GA uniques is not the same thing as Compete uniques is not the same thing as server log uniques. Even if GA, Compete and server logs had recorded all of the traffic to a site, all three would still show a different number uniques do to the different ways uniques are defined and determined.

    The most important things to take away from this thread is that for those who need a third-party overview of other sites traffic that Compete is a very good source and that there is absolutely no excuse for still using Alexa.
    Ken Barbalace - EnvironmentalChemistry.com (Environmental Careers, Blog)
    InternetSAR.org: Volunteers Assisting Search and Rescue via the Internet
    My Firefox Theme Classic Compact: Based onFirefox's classic theme but uses much less window space

  5. #35
    Registered
    Join Date
    Aug 2006
    Location
    Sacramento, CA
    Posts
    208
    Quote Originally Posted by KLB View Post
    But if that data is skewed, it doesn't matter how many data points you pick. the results will still be skewed. There is no way around it.
    So if Alexa toolbars were installed on every computer they would still have a webmaster skew? Thats what you are saying.

    Quote Originally Posted by KLB View Post
    NO, that is what you may be reading into what I said, but it is not what I'm saying. What I'm saying is that Compete is the most accurate indirect traffic reporting service I have seen. Notice the key word "indirect". GA and server logs are direct traffic reporting tools.

    I'm also trying to get people to realize that GA uniques is not the same thing as Compete uniques is not the same thing as server log uniques. Even if GA, Compete and server logs had recorded all of the traffic to a site, all three would still show a different number uniques do to the different ways uniques are defined and determined.

    The most important things to take away from this thread is that for those who need a third-party overview of other sites traffic that Compete is a very good source and that there is absolutely no excuse for still using Alexa.
    I agree Compete is more accurate for most comparisons, especially across different niches. I never said it wasn't. I'm just saying it is not a reason to completely ignore data from Alexa. Both can provide useful information depending on what you are comparing/measuring.

    The only thing I am disagreeing with, is when you said the difference in my Compete numbers and Analytics numbers could be explained due to counting differences. A unique is a unique, the difference is simply explained by the fact that the <1% of the internet users that Compete samples use my site about half as much as the entire internet.
    ________
    UPSKIRT TENNIS
    Last edited by rpanella; 03-17-2011 at 10:47 AM.

  6. #36
    Site Contributor KLB's Avatar
    Join Date
    Feb 2006
    Location
    Saco Maine
    Posts
    1,181
    Quote Originally Posted by rpanella View Post
    So if Alexa toolbars were installed on every computer they would still have a webmaster skew? Thats what you are saying.
    No that is not what I was saying because I wasn't discussing Alexa in my comparison. I was comparing Compete, GA and server logs. If Alexa toolbar was installed on every computer the skew would be eliminated but the data still wouldn't agree with the other three because of different definitions of "uniques". What is a "unique"? How do you determine this?

    Quote Originally Posted by rpanella View Post
    I agree Compete is more accurate for most comparisons, especially across different niches. I never said it wasn't. I'm just saying it is not a reason to completely ignore data from Alexa. Both can provide useful information depending on what you are comparing/measuring.
    Alexa should have been totally ignored for years now. Its data is worse than worthless for the following reasons: 1) some spyware applications label it as spyware and remove it; 2) many IT department in academic institutions, government and corporations ban the install of Alexa; 3) most regular users never install it on their own; and 4) webmaster/SEO types typically do install Alexa. Those four factors cause Alexa's sample to be extremely skewed and way too small to be statistically relevant. The only one who gains from Alexa is Amazon.com in that it theoretically improves their ability to target book and purchase suggestions when the person goes to Amazon's website.

    Quote Originally Posted by rpanella View Post
    A unique is a unique, the difference is simply explained by the fact that the <1% of the internet users that Compete samples use my site about half as much as the entire internet.
    Again, define unique and explain how a unique would be calculated. How do you ensure that automated processes like site scrapers/bots/spiders/etc. don't get counted as uniques?
    Ken Barbalace - EnvironmentalChemistry.com (Environmental Careers, Blog)
    InternetSAR.org: Volunteers Assisting Search and Rescue via the Internet
    My Firefox Theme Classic Compact: Based onFirefox's classic theme but uses much less window space

  7. #37
    Registered
    Join Date
    Aug 2006
    Location
    Sacramento, CA
    Posts
    208
    Regardless of how you define unique visitors, using direct metrics will give you a very accurate count of the true number of unique people that come to your site. These would include Google Analytics or various server stat packages, ad network stats, etc.

    Regardless of how you define unique visitors, indirect metrics such as Alexa, Compete, and Quantcast can only "guess" how much traffic you get based on a very small sample of statistics they collect.

    You have an extreme bias against Alexa, thats fine, no one is forcing you to use it. I'm simply saying you cannot compare a direct metric such as Analytics to an indirect one such as Compete, and explain the difference in numbers due to different definitions of what a unique is. It has nothing to do with that.

    I've wasted enough time trying to explain this to you, and you continue to go on tangents bashing Alexa or comparing counting uniques with IPs or cookies, which is not what this thread or I was ever talking about.
    ________
    Yamaha yz125
    Last edited by rpanella; 03-17-2011 at 10:47 AM.

  8. #38
    I see mildly ill people. AmbulanceBlues's Avatar
    Join Date
    Aug 2006
    Location
    Houston
    Posts
    119
    Not to step into somebody else's argument, but...

    I was intrigued about the "Big Brother" aspect of how they get their clicks. I went to the Compete FAQ and it does an excellent job of using a whole lot of words to tell you almost nothing about that subject.

    I seems to me that when they refer to getting click information from ISPs, that could mean one of two things (actually, I guess it could be both or neither):
    1. They buy the logs of an ISP's customers' activities OR
    2. They buy the logs of random internet traffic going through an ISP's Internet backbone exchange.

    The first I would have a serious problem with, because a company with whom an account that I pay for should assume that any data I may generate is a part of my account's activities. To me, that's just like my bank selling "randomized" transaction information. It would be wrong.

    The latter, however, I don't think I really have a problem with. The nature of the internet means that your web surfing is going to be routed through numerous exchanges owned by entities that owe you no privacy. If I operate an exchange on the internet backbone, I'm accepting every signal that comes to me without condition and routing them appropriately without fealty to anyone but my own clients who benefit from the service. In that situation, I wouldn't have any compunction about selling logs that don't directly relate to my clients or anybody else who doesn't have a reasonable expectation of privacy.

    It seems compete would want to go with some version of that mechanism because it's also less geographically limited. If they were buying surfing information about an ISP's customers it would necessarily be limited to that ISP's customers. (Duh...) What I mean to say is that what if all of their "Two Million Users" are customers of ISPs in the San Francisco Bay area, or the Houston area, or any other limited (or dispersed but patchy) area?

    The "Two Million Users" statement does kind of put a kink in my theory. But if they're doing what I think they're doing, I don't see any privacy problems there. You don't have a reasonable expectation that information you send over the internet (without efforts made to secure it via tunneling and/or encryption or whatever) will remain private. If the ISPs are doing that, it seems to me they're only monetizing an opportunity that presented itself.

    EDIT:

    Okay, I went back and re-read the articles in the first post and it appears I am wrong. When I first perused them it sounded like they were just making conjecture like me but I was wrong. But I still think my idea is better (and more ethical) than what Compete doing. When they rip me off, remember you heard it here first.
    Last edited by AmbulanceBlues; 03-25-2007 at 11:37 PM. Reason: I'm stupid.
    --> --> --> --> --> -->

  9. #39
    Registered
    Join Date
    Aug 2006
    Location
    Sacramento, CA
    Posts
    208
    Very good point AmbulanceBlues, data from a backbone provider would be more ethical and much more random.

    Data from a particular ISP alone can be very biased. Most ISP have a start page which search powered by either Google or Yahoo. My site is basically banned from Yahoo, and 85% of my traffic comes from Google and the rest is links or typein. You think my site would be fairly represented taking data from all the Yahoo SBC dsl users? Not at all, but an ISP powered by Google would probably over estimate my traffic. Also, dialup users are overall probably less tech savvy, so those numbers would undercount stats to tech sites. You would need to buy random stats from all ISPs to get rid of the bias.
    ________
    Buy roor bongs
    Last edited by rpanella; 03-17-2011 at 10:47 AM.

  10. #40
    Site Contributor KLB's Avatar
    Join Date
    Feb 2006
    Location
    Saco Maine
    Posts
    1,181
    AmbulanceBlues has touched on an aspect of this issue that I tried to touch on and has not been well discussed.

    Rpanella you are absolutely correct that getting data from only a few ISPs would be skewed and I touch on this in post #10

    The backbone idea for collecting random data would be good, however, it would still need to be the backbones from several providers AND it would miss data that never made it to a backbone. For instance if the website and the user were on the same ISP those requests may never get routed to a backbone.

    I suspect that Compete is buying data that is culled from ISP logs. The question is how detailed this data is. I do suspect a serious potential for privacy issues if this is not handled properly (ala AOL search log release).
    Ken Barbalace - EnvironmentalChemistry.com (Environmental Careers, Blog)
    InternetSAR.org: Volunteers Assisting Search and Rescue via the Internet
    My Firefox Theme Classic Compact: Based onFirefox's classic theme but uses much less window space

  11. #41
    Chronic Entrepreneur
    Join Date
    Nov 2003
    Location
    Tulsa, Oklahoma, USA
    Posts
    1,112
    compete.com is showing less than 1/3 of the unique US visitors that Google Analytics is reporting for me. Compete seems better than Alexa for comparing relative traffic across multiple sites, but they seem pretty worthless as far as reporting actual visitor numbers.

  12. #42
    Site Contributor KLB's Avatar
    Join Date
    Feb 2006
    Location
    Saco Maine
    Posts
    1,181
    Quote Originally Posted by Westech View Post
    compete.com is showing less than 1/3 of the unique US visitors that Google Analytics is reporting for me. Compete seems better than Alexa for comparing relative traffic across multiple sites, but they seem pretty worthless as far as reporting actual visitor numbers.
    I'll have to look closer at the "unique" traffic comparisons between GA and Compete for my site.

    Unlike Alexa, Compete might simply be reporting the actual traffic it sees rather than trying to extrapolate "real" traffic numbers. If this is true then it would only be useful for comparing one site against another, but not for getting actual traffic levels of a specific site.

    For me what I really care about with my own traffic is pageviews and visits per day and total pageviews per visit. I don't really care about total unique visitors per month because I see this as a totally fictitious number regardless of methodology (nobody has given me a definition of "uniques" yet).
    Ken Barbalace - EnvironmentalChemistry.com (Environmental Careers, Blog)
    InternetSAR.org: Volunteers Assisting Search and Rescue via the Internet
    My Firefox Theme Classic Compact: Based onFirefox's classic theme but uses much less window space

  13. #43
    Chronic Entrepreneur
    Join Date
    Nov 2003
    Location
    Tulsa, Oklahoma, USA
    Posts
    1,112
    Quote Originally Posted by KLB View Post
    For me what I really care about with my own traffic is pageviews and visits per day and total pageviews per visit. I don't really care about total unique visitors per month because I see this as a totally fictitious number regardless of methodology (nobody has given me a definition of "uniques" yet).
    What I'm refering to as "uniques" here is monthly unique visitors -- how many unique people have visited the site within a one month period. The same person visiting the site multiple times during the one month period is only counted once. Of course, even with this specific definition the numbers will vary somewhat depending on how unique visitors are tracked (cookies or IP's... javascript or images...etc.)

    Absolute monthly uniques is the main metric that most big ad buyers want to know when evaluating direct ad buys. Their main concern seems to be how many unique eyeballs will be seeing their ad.

  14. #44
    Site Contributor KLB's Avatar
    Join Date
    Feb 2006
    Location
    Saco Maine
    Posts
    1,181
    Quote Originally Posted by Westech View Post
    What I'm refering to as "uniques" here is monthly unique visitors -- how many unique people have visited the site within a one month period.
    Great, but how do you determine this? No method can truly track this. People share computers, IP addresses, browser profiles, etc. People also use different computers, connect from different locations, use different browsers, etc. Cookies get deleted, IP addresses change, etc. Quite simply the concept of "unique" visitors per month is a fraud. At best you can determine unique visitor's per day.

    For example if a teacher uses a site like mine in her class, she might teach the same subject to three or four classes. These students would share classroom computers thus multiple "uniques" would actually be sharing IP address and cookies. If a user connects via dialup, each time they could get a different IP address or share the same address with thousands of other users (e.g. AOL). Thus you can't rely on IP address to determine uniques. Many users could use the same computers in an Internet cafe or library and thus share the same cookies.

    How do you deal with bots and spiders that spoof regular browsers? More and more of these bots are part of zombie networks that infect regular computers and thus scrape pages from multiple "regular" IP addresses, thus non-human visitors are becoming nearly impossible to separate from human visitors.
    Ken Barbalace - EnvironmentalChemistry.com (Environmental Careers, Blog)
    InternetSAR.org: Volunteers Assisting Search and Rescue via the Internet
    My Firefox Theme Classic Compact: Based onFirefox's classic theme but uses much less window space

  15. #45
    Administrator Chris's Avatar
    Join Date
    Feb 2003
    Location
    East Lansing, MI USA
    Posts
    7,049
    Great, but how do you determine this? No method can truly track this. People share computers, IP addresses, browser profiles, etc. People also use different computers, connect from different locations, use different browsers, etc. Cookies get deleted, IP addresses change, etc. Quite simply the concept of "unique" visitors per month is a fraud. At best you can determine unique visitor's per day.
    Daily uniques aren't immune to those same issues. The same percentage of a whole should be affected and thuse the same margin of error.
    Chris Beasley - My Guide to Building a Successful Website[size=1]
    Content Sites: ABCDFGHIJKLMNOP|Forums: ABCD EF|Ecommerce: Swords Knives

Similar Threads

  1. Alexa
    By izwar in forum General Promotion
    Replies: 12
    Last Post: 04-09-2007, 09:46 PM
  2. Alexa shows networks of sites from owner
    By Tijmen in forum Internet Industry News
    Replies: 6
    Last Post: 03-16-2006, 12:49 PM
  3. Grabbing a sites' Alexa rank
    By Masetek in forum Website Programming & Databases
    Replies: 2
    Last Post: 11-28-2005, 06:41 AM
  4. Banning Alexa Robot
    By Blue Cat Buxton in forum General Management Issues
    Replies: 1
    Last Post: 08-17-2004, 10:42 AM
  5. Alexa Amazon Associates?
    By RockNRollPig in forum Advertising & Affiliate Programs
    Replies: 2
    Last Post: 02-20-2004, 02:42 AM

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
  •