Slashdot Mirror
Alexa, Amazon's Most Flawed Idea
Rub3X writes "The Alexa ranking system is naturally flawed. The data should never be treated as accurate, as it's easily manipulated, and not supported for most browsers in the world. It's an estimate, and nothing more.
" I've been saying that forever, but unfortunately for me, since it's a number on a website that is considered "Real" to some, I'm supposed to take it seriously. I imagine this is a problem for many webmasters out there.
I would like to append to Taco's first salvo that "everything else Digg users like sucks too."
Yes, this is tongue in cheek. Mod me troll anyway.
http://www.alexa.com/data/details/traffic_details? &range=&size=large&compare_sites=&y=r&url=ww.com
:)
not on an absolute scale, but you can compare trends, and if YOU don't fake the data you're ok
remove space for link to work
Services like Megaupload.com force Non-American/Non-european users to install Alexa toolbar to download the file.
That explains why Alexa has file-upload sites such as Megaupload,rapidshare in the top 10 sites of most countries...
Wincopy
According to the article:
:P
"Alexa has no support for FireFox, Opera or Safari at all. "
According to Alexa's Wiki:
"Users running any browser except Internet Explorer and Mozilla Firefox are not represented. Thus users of Opera, Safari, mobile phone (WAP) browsers are all ignored. Nevertheless, this is still the vast majority of the browser market."
So its half right
I've pointed this out before. There are weird statistical anomolies that should show that Alexa's webratings are not perfect. Take a look at this data for Slashdot and Digg. The traffic ratings both shoot up withing a s short amount of time. It just doesn't make much sense. http://www.alexa.com/data/details/traffic_details? &range=2y&size=medium&compare_sites=www.digg.com&y =r&url=www.slashdot.org#top
Ooo man the floppy drive is broken. No wait. The computer is just upside down.
but this is sort of a non-issue. sorry
Alexa may not be supported by most browsers, but it is supported by The One Browser that most people use and almost everyone has.
I remember for a while LewRockwell.com, which promoted alexa for its readers, was top-500, beating out worldnetdaily.com and gamefaqs.com. Now, nothing against LewRockwell.com, and it is indeed surprisingly popular, but there's no way in hell it's a top 500 site.
Apology to Ubuntu forum.
Everyone who owns or develops web sites knows this. Anyone who hints in a forum the numbers may be accurate immediately gets slapped down. It's the non-technical advertisers who don't know this. And they're the only ones who care about this ranking in order to gauge how much to spend on purchasing web site advertising. Since almost no web sites publicly display traffic info advertisers find Alexa rankings very convenient and probably just don't understand why they'd be useless.
Until advertisers "get it" or a much more accurate public metric is made available, Alexa rankings will unfortunately matter to web sites that are supported by advertising.
Developers: We can use your help.
Last time I checked, the term 'most' meant a majority.
Firefox, Safari and Opera may have significant market penetration, but 30% a majority does not make.
Now for my slashdot rant: I remember when slashdot used to post news. You know, 'news for nerds. stuff that matters'? Lately, the tagline might as well change to Slashdot: "Some idiot posted this somewhere on the web. We'll ride their coat tails."
These days, slashdot hardly has as much credit as celebrity gossip sites such as What Would Tyler Durden Do? and The Superficial. At least they're honest about the fact that they are posting little more than rumors.
I currently have no clever signature witicism to add here.
Alexa also doesn't calculate masked domains.
I have a blogger blog that is masked with my own domain name.
Yell & scream & rant & rave... it's no use... you need a shaaaave ~ Bugs Bunny
Yep - Alexa's sample size is pretty small, but skewed toward IE folks and people who install/game it so their web sites rank well ... although I think anything in the top-1000 is almost always "something" significant. Unfortunately, there aren't a lotta metrics out there (plus those often provide varying results), and this is easy to understand and free, so it's often used.
Best source is the source - i.e. would be real interesting to know what the web stats (for actual web logs) are like for a site like Slashdot - I can only imagine the number of hits/page-views/etc.
Hulk SMASH Celiac Disease
> not supported for most browsers in the world
You heard it here first, folks: IE and Firefox make up only a small minority of web browsers in use.
Star-based rating systems are useless for more than getting a quick idea of what's up. They don't really tell you anything; for instance, I've purchased items in the past that have issues that don't bother me that I would have passed on just based on a "star" approach.
This goes for Alexa, this goes for movies, etc. I suspect that most consumers of this sort of information use it like I do -- only as a starting point to filter out the really bad products. For anything important or where I'm spending more than a few bucks, I'll read the reviews of a product as it's still the only way to really get any good information.
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
..even though (1) most people in the world do use Internet Explorer (2) there are Alexa extensions for Firefox, etc. (3) Google does the same sort of traffic aggregation in their toolbar, under advanced options when you install. I mean, if Amazon does it: bad; if Google does it: awesome. Right? All these people talk about how easy it is to manipulate TrafficRank, but can anyone actually prove --- with real numbers, not empty assertions --- that they've done it?
The problem is that statisically it's nice to say that 30% does not make a majority but Im sure that spreads changes from website to website. Imagine looking at the statistics for a Linux website. The majority there better not be IE.
Ooo man the floppy drive is broken. No wait. The computer is just upside down.
Alexa also rewards webmasters who write badly broken IE only webpages, forcing people who normally use Firefox to switch over to IE for that webpage.
I read the internet for the articles.
Now it's clear that the rankings from this system are heavily skewed and misses a substantial portion of the user base.
This suggests it is useless as a way to estimate how much to pay for advertising on a web site (though since this is usually per click/per display I don't see why ranking matters here). However, it doesn't show that this data can't be usefull for other things. For instance it could be quite usefull to know what other sites the users (or IE users) of a site visit.
In other words the data seems useless for any statistical analysis but it could be quite helpful to know what sorts of users visit a site. Sure slashdot's traffic might be underrepresented but I bet you the data still show that slashdot users are quite likely to go browse gadget purchase sites or programming related sites. If you want to know where to advertise your new fancy gadget or a fancy new programming enviornment that would be very usefull information even if it wouldn't support a rigorous statistical analysis.
If you liked this thought maybe you would find my blog nice too:
The SearchStatus plugin for Firefox displays the Alexa rank and the Google Pagerank for a page graphically side-by-side. I use these to get a quick idea about the popularity of a site. The two numbers are a lot more useful together. For example, PageRank Zero and a high Alexa Rank implies that the site is either brand new or trying to cheat the ranks.
One fact TFA and the Slashdot title both got wrong, is Alexa wasn't Amazon's idea. Until Amazon bought it in 1999, Alexa was the commercial offshoot of archive.org for three years. Alexa is still what gives the Wayback Machine its web crawls.
Slashdot Burying Stories About Slashdot Media Owned
Not only is the idea flawed (unless they get a lot more users), but the metric by which they claim to rank is not properly implemented. They claim to multiply the reach and page view numbers (well, they say it's the geometric mean, but the ordering is the same), but reach clearly gets weighted higher. I've compared their numbers against their rankings and found this to be true in all cases.
Given the slashdot's geek crowd, its hard to grasp that alexa gets its numbers from Alexa toolbar installed slashdotters..
So a corollary of that would mean that,higher the number in Alexa, higher the number of 'lame' users of a website who actually installed a Alexa toolbar.
Wincopy
Kevin Smith uses his message board to talk with his fan base, and a few years ago he had an interesting "discussion" regarding Atlanta (now-Boston) DJ Fred Toucher over the film Jersey Girl.
In comparing his website's Alexa numbers to the numbers for Toucher's website, Kevin quotes numbers for both sites, followed by the line "Yeah - I'm not a success" after indicating how many more Alexa users visit his site over Toucher's.
If Alexa is good enough for Kevin Smith, it should be good enough for the rest of us, right?
For Firefox, you can use the SearchStatus extension (download from either the Firefox add-ons page or their home page). It's actually a somewhat useful tool also, in that it displays Google PageRank and Alexa rank for each site you visit, and has a few decent tools for showing various search engine related information for a given page. It also feeds data on every page you visit to Alexa as a byproduct of looking up their Alexa rank, which may be a positive or a negative for you. I personally verified that Alexa rankings change as a result of this Firefox extension, based on the fact that a couple of my personal pages (which I generally look at several times a day) were unranked prior to me installing the extension and then had a large spike in traffic not long after I installed it.
I pointed out to my boss awhile back when he was complaining about our Alexa rank that if he actually wanted the rank to improve, probably the easiest way to do it would be to have every person in the company install and use either the Alexa toolbar for IE or the SearchStatus extension for Firefox. If you're not one of the top hundred sites (or so), then Alexa ranks seem pretty easy to manipulate. Having company employees install their toolbar isn't even gaming the system per se, it's just making sure that people who are likely to be visiting the sites you care about are "well represented". As for why people put so much stock in Alexa rankings despite the obvious facts against their reliability, it's simply because there's nothing better out there. I'm sure Google could do what Alexa does much better if they felt like it, based on both search engine traffic and the Google toolbar users (has to be a LOT more of those than Alexa toolbar users). Then marketing drones would be watching those ranks obsessively instead. They take what they can get.
Unfortunately, it doesn't matter if it should be trusted, it is in any case. I've even seen Wikipedia use Alexa rankings as a basis for whether a website is notable enough for its own article...
But I guess statistics have always been used to allow people to fool around with fantasy and avoid facing reality.
it's a number on a website that is considered "Real" to some
That's not real, that's int.
Send email from the afterlife! Write your e-will at Dead Man's Switch.
Alexa doesn't rank itself but you can chek their stats at adbrite site http://site.www.adbrite.com/mb/commerce/purchase_f orm.php?opid=19607&afsid=1&spid=15
they get just over 5000 visitors a day, I don't know how many people installed their toolbar but they must be multiplying visitor numbers and traffic stats like crazy. With such a low user base the ranking stats they have can't be very accurate.
I have couple of sites and on that gets twice more visotrs, according to my server logs, is ranked way lower on Alexa the the other.
sweet
If by some chance you did get the first post, would you really want to wasting by saying that?
nothing
Just did a quick non-scientific test on my webserver for my friends website. He runs a page for his punk band and links some stuff in his myspace page. Let's see how alexa did
[root@### logs]# cat access_log | grep 200 | grep -i alexa | wc
33 883 9536
[root@### logs]# cat access_log | grep 200 | grep -i mozilla | wc
46729 1024301 11454746
One in 10 my ass
Correction: would you really want to waste it by saying that
nothing
http://paulgraham.infogami.com/blog/alexadanger
Alexa is flawed from the start.
What impetus or benefit would a user have to install a toolbar that tracks them? Other than out of charity to help out this company? I don't get it. Nor do I particularly trust them. Just one more thing to help crash IE.
Your ideas are intriguing to me and I wish to subscribe to your newsletter.
Nobody really denies that Alexa is a bit crap anymore. It's so easy to manipulate you'll wake up one morning and Joe-Bob's myspace page will appear to be the most visited site on the Internets. The problem is, that nobody has really offered a competing solution.
A solution that allows you to track visits to any given website must be something on the user end. You can't expect every website to install some piece of tracking code. It might be possible if a service like Alexa was standardized and put into all browsers from the get-go, but this brings up privacy implications and would never happen in any version of reality.
Google has it's own system for determining the importance of a page - and while it's still flawed, and only really geared towards their own goals, it does a good job of showing the importance of a website. Rather than ranked #1-infinity, though, every page is ranked from 0 to 10. Not as specific, but about as useful a ranking as you're ever going to need.
or else!
Me thinks the author has a bone up his buttocks.
1) IE is, love it or hate it, the dominate browser by a huge margin.
2) Most users do change their home page, but how is it completely invalid that default setting users hits on MSN (his example) don't count? They went to the page, likely did view something, they generated traffic and likely impressions if not click throughs (Though the odds of that are higher than hitting some random page from a google search query and paid placement result).
3) What alternative exists today in full, working form, that supplants Alexa and it's real usefulness to people? None? That's right... none. So put up or shut up.
4) Alexa numbers are useful... not entirely accurate, but useful at least, in valuing traffic and promoting sites or working with less than computer literate PR folks. They are not entirely innacurate either and certainly are a good indicator, especially for site traffic related to IT/IS oriented sites (And I mean a lot of different markets encompassing a lot of sites).
Marketing people in agencies and media companies haven't heard of Alexa. The first time I checked it out was after a support technician told me about it. Every article I read and every salesperson I talk to mentions either Comscore or Neilsen data. Alexa is sketchy, but that doesn't matter, because nobody who buys and sells ads cares.
Seriously, why bother writing two or three sentences anymore? Just put a single link on a single word, that's even less helpful and even less work for the editors.
WTF IS ALEXA?
Another case of "I don't want to waste 30 seconds to explain WTF the news is about, let 50K users waste a few minutes and slashdot a website trying to figure out what it is".
This Alexa thing sounds like a great tool for anyone wanting to know which websites are most frequented by IE users who are susceptible to Internet advertising.
--- Tao
Whenever you conduct a poll, and that's what Alexa is doing, you are always excluding data from those who do not respond to polls (for whatever reason). It's an inherent flaw in polling.
Web 2.0 == Giant Blogspam Circle Jerk
Wikipedia constantly uses Alexa to see if linking to a website or profileing a website is "notable". Despite outrage by the people who submitted the content, usually everything that gets nominated for deletion has some editor cite alexa as a reason to delete it.
This is great for /. stats and Alexa. IE users just entered a page view to find out.
Why would anybody want to install their toolbar when they are known to be a spyware company?
My Web site has a Google PageRank of 7, but an Alexa rank of 261,144. I can't be the only one that has a site with good GoogleJuice and a good number of visitors, but an Alexa rank that falls below Jean Teasdale-esque Geocities sites with angel and "survivor" glurge.
Alexa rankings will always be worthless compared to the site traffic logs, but it does have some uses. It's a "big picture" tool at best and can be used to spot trends in traffic growth / decay. When I work on linking strategies or affiliate marketing I use Alexa data in a general way to drill down the huge pool of possible targets. It enables me to sort a long list easily.
Another thing it's good for is during campaigns. The spikes in traffic during a promotion can help give an idea of its success. Obviously not accurate, but a significator of trends. It usually reflects fairly accurately what I find in the traffic logs themselves - for instance a jump in traffic for October during the "xxx" promotion or whatever....
What's important is that all Web analytics and sampling have flaws. There is no perfect tool. Agencies use data from Statmarket, Netratings, Onestat, etc every single day to make business decisions, even when this data varies widely.
My own method is to gather what I can from tools like Alexa and compare it with others. Contrast, crunch, find something useful. Hopefully.
Doesn't make the stats any more accurate but at least it makes them look pretty.
One of the things that most Internet marketers miss (myself included at times) is that as you move up the food chain, there are more buyers. If I am pushing a brand-less product, I focus on my CPC rate, conversion rate, etc., and don't care where the ad runs, only the conversion rate there and what I am paying per click.
However, the big boys (Ford Motor Company, Warner Bros. Television, etc.) focus on brand building and budgets. They don't ask if they are making money off the impressions, they have a quarterly budget, spend the budget, and ideally aim for the biggest bang-for-the-buck. However, they don't want their brand associated with marginal content, so they have to approve the site carrying their ad.
Now, they pay dearly for the privilege, but for a content site, a big-name ad agency placing a big brand may be willing to pay substantially more on a CPM basis because they need to cut a deal, take up time, etc. If I am making $5 CPM, I'm not going to spend hours with their legal department, etc., to make $5.05 CPM. Now for $6 or $7 CPM, or even $10, I'll spend time handling them as a client, not running an ad network.
The flip side to this, the ad agency doesn't want to spend hours time from highly paid professionals for a site with 1000-2000 visits/day. They need a bang for the buck, and if their time is going to add $5-$10 on a CPM basis for compliance, then the advertising becomes less viable.
So, when you hit thresholds on major rankings, whether it be Alexa, Nielson, etc., then some of the ad players notice you. So the rankings matter, because maybe hitting top-2000 means you can sign on a few premier accounts. While the traffic jump from spot 2001 to 2000 may not be huge, you may see a HUGE revenue jump if you can sell the inventory.
That's why the things matter, as you move up, companies that won't deal with smaller players because transaction costs are too high show up... More demand for your ad inventory, holding supply constant, means higher prices. Even as you move higher, the supply increases (which would force you to take lower prices to move it), more buyers come on the scene, which usually means a still higher CPM.
The rules change as you move up the food chain, which is why the people that define the food chain matter.
Alex
I refer you to the "plog".
the major advances in civilization are processes which all but wreck the societies in which they occur - A.N. White
with the penetration of google toolbar they could have solved this long time ago. alexa toolbar tracking is just not so dependable as google tracking could have been. but i guess that would only cast light on the dirty truth of much lower exposure web has thus hurting google s adsense program...
Alexa's usefulness is more statistical than exact. I don't use it to see specifically how much traffic a site gets, but to see how it changes over time. As long as their measurement standards don't deviate too much, then this can still be useful information.
When Alexa first came out, I was willing to use it. There were two features that it provided, and page ranking was actually the least important. Far more important to me was the goal of building an inverted index of the web -- tracking who linked to this site I was looking at, rather than seeing who this site links to.
:-).
All that changed when Alexa was bought by Amazon. And then the truth came out -- all the information that I thought was private was in the database, and now owned by a commercial company, with no restriction on how they used that information. All the information about me that came to the right of the question mark was now in a commercial database, just as bad as AOL's release of search engine queries.
That gave a 100% loss of trust for me. And not just me.
People who know what's going on won't install Alexa because it's giving unrestricted access to personal information to a commercial company for their own profit. And, the "backwards index" -- which helps the internet navigation globally -- is no longer the focus of the product.
So for most people, it has lost any purpose and functionality.
This is why it is so fundamentally off on any numbers it generates. Heck, Neilson ratings have to be more accurate
Sorry but Alexa is still useful to guage trends and "generalized" site popularity.
Here's a great example: POXNORA STATS
PoxNora is a game that was slashdotted last week. See the big spike in their traffic graph (roughly Oct 13/14)?? That's when they got slashdotted. Don't tell me Alexa stats are completely useless.
Any webmaster worth their weight in salt knows Alexa data is flawed. This isn't news for nerds.
I do use alexa to measure the relative worth of my sites vs competitors. The data never conicides with my own analytics against referral logs and that's ok. Alexa sucks like that.
How is it that Anybody chooses to run the Alexa toolbar, when "tracking which web sites you visit" is such a clear synonym to "Spyware"? This reveals another skew on the ratings that Alexa can provide: Any site that gets a high Alexa rating is a site that is frequented by Inexperienced and gullible Internet Explorer for Windows users who don't mind having spyware installed on their computer. This seems to me to mean: If the site has a high Alexa rating, it is a site that I don't want to visit. Alexa is... Spy Ware!!
What is the reason for web stats? If you're paying per view or per click then the information is directly available.
This leads to an interesting possibility. The ad providers could provide a ranking of sites based on the number of adds that they show there and the number of clicks that are created. This is, of course, open to manipulation via click fraud and other techniques but it would probably be more accurate than Alexa's rankings.
Then, if you wanted to improve this even more you could combine this with the number of searches that go to a page. A large net firm that provided these services could do such a ranking. Google or Yahoo could do this. Perhaps they do, for their internal consumption.
If you have the alexa tool bar and you check your website a few times a day. Your rating will improve by 100,000's. =) My traffic hasn't changed for www.housefox.ca at all in the past year but it has gone from an alexa ranking of 200,000 to 580,000 in a period of 6 months because I stopped checking it daily own my computer with the alexa bar installed.
Check it out for yourself.
Anyone notice Alexa is not ranked?
? url=alexa.com
http://www.alexa.com/data/details/traffic_details