Here's Why I Don't Trust Alexa Rankings
A spirited debate has been gong on recently around hits, visits and unique visitors. In the comment stream of Real Estate Blog Traffic, How Many Hits Should I Expect, Brad Carroll made the following comment.
"Does one blog get more traffic than another? Simply go to alexa.com and find out. Granted, it's not perfect and there are ways to "work the system" but it gives a great overview.
Brad certainly acknowledges it is not perfect and we both commented back and forth about this, but I felt an illustration of why I don't trust their numbers was in order. Especially since I think Brad is one of the smartest SEO guys I've ever read. So, this is not a knock on Brad in any way, shape or form. I know he's going to enjoy this comparison.
In my comments, I suggested we use a few real numbers and comparisons to test it out. So, I went to the real estate segment of topblogsites at Blogflux and pulled three sites from their rankings. I used TopBlogSites because in order to be on their rankings, you must insert code that sends messages to their data collection servers. This is a very accurate way to measure one site against another. There is no guess work.
The Three Blogs I Chose To Compare
I chose the number 1 ranked site, HousingPanic, the number 12 ranked site, BloodHoundBlog, and the number 43 ranked site, RSS Pieces. The one measure of true traffic, for me, is unique visitors. This is a measure of traffic that weeds out multiple visits by the same "person" using IP addresses. While visits is certainly important, unique visitors will more accurately measure the true reach of your blog. I could get 50 visits from my mom in one day, so I discount visits.
A Direct Comparison Of Unique Visitors For This Month / All Time
- Housing Panic - 65,931 / 968,694
- BloodhoundBlog - 8,985 / 176,033
- RSS Pieces - 2,216 / 48,073
As you can see, there is a large difference between the three blogs actual unique visitor counts. I have compared TopBlogSite's numbers with our own number at the Real Estate Shows Blog and they are right on target. So I have some degree of confidence in this comparison.
So, if Alexa is a good way to judge a blogs traffic, you would expect that the Alexa rankings would reflect the numbers above. But they do not. I wish I could embed the live graph of the Alexa traffic rank comparison, but AR doesn't allow Java embeds, so I have to embed this screen capture.

As you can see, for much of the last three months, RSS Pieces ranked much higher than BloodhoundBlog. HousingPanic doesn't even show up on the radar. HousingPanic has had more than 20 times the visits of RSS Pieces and more than 5.5 times the traffic of BloodhoundBlog, and it doesn't even show up in the Alexa comparison. Their traffic rank right now is 154,799. BloodhoundBlog's is 71,704 and RSS Pieces' is 41,890.
Alexa's Ranking Of These Three Blogs Is Completely The Opposite Of Their Unique Visitor Rankings.
This is why I don't trust Alexa's rankings.
UPDATE:
TechCrunch also wrote about Alexa today. They were even more harsh: http://www.techcrunch.com/2007/08/13/alexa-says-youtube-is-now-bigger-than-google-theyre-wrong/
Jeff Turner aka respres

I can be ranked number 5 million in alexa but what is that to me if i make more than the person ranking number 50,000? I just dont get why people love alexa so much ;-)
Michele... thanks. I think it's a fair comparison.
Mark... as an example, I don't have the Alexa toolbar installed on my browswers. So, Alexa can't see my visits to my own sites.
Geez. If there are "ways to work the system", what good is the system.
Bet a Dollar to a Donut that if I installed Alexa on Homefinders.com, my ranking would improve.
Wouldn't bring any more visitors.
Lenn... 100% accurate in my opinion.
Susan... I'm glad you stayed, despite learning that I'm not Tim McGraw. :)
Jeff: I had installed Alexa on our computers several years ago and enjoyed watching the improved rankings of our two sites. What I did not discover until later is all the spyware that follows the Alexa download.
Ultimately, I came to the same conclusion about Alexa. Statistically, it's probably only valid for the top 1000 or so sites.
Roberta... my ONLY point in writing this was to illustrate Alexa inaccuracies. As you state, the reason for this is the way they collect data. It makes sense that RSS Pieces would rank higher, since their blog is read by a lot of tech types who pay attention to SEO. The Alexa tool bar is used by tech types. This was not meant to disparage anyone, just to highlight that Alexa is not the place to go to get a clear picture of visits to your blog.
Sharon... I don't have any direct knowledge of spyware, since I've never installed their software.
Diane... RSS Pieces builds great Blogs that have great SEO optimization. In my opinion, they are among the best in the business. I don't have a background piece on this. The only goal of this post was to point out Alexa inconsistencies.
Greg... checking and double checking is the best way to get a clear picture, though none of the ranking services, including TopBlogSites is free from error.
Jeff,
Thanks so much for posting this. I have been right in the middle of this topic and for some reason always held Alexa to an undeserved level of superiority... yet could never figure out why I wasn't getting the statistic results I anticipated. It has baffled me forever. :P
The 4 tracking systems that I have found to bring the most consistent results are Mint, Google Analytics, SiteMeter and BlogTopSites/BlogFlux. Their results are always within about a 2-3% margin of each other, and I live all four for different analytical reasons.
Jim... yeah. Alexa can be fooled pretty easily. Simple enough to just have a all of your clients install the toolbar and come visit your site a lot. It won't actually do anything for you, but it will make your Alexa numbers look better.
In the end, analyzing server logs is the only way to really get at "truth." We use a very high end analytics package to monitor and extract information from our server logs. The only reason I use TopBlogSite is so others can compare us on a somewhat level playing field. There are flaws in all of the online measure tools, but together they'll paint a decent enough picture.
Maureen... the right people reading your blog is more important the lots of unique visitors that aren't your target clients. That's the bottom line.
Ginger... what ultimately counts is that your blog is generating readers who turn into clients.
Okay I am mad. First you steal my CSS code and then you discredit me! HEHE No. I am only kidding. Great Post. I think you hit the nail on the head about "tech types" skewing the numbers. I had a feeling that the results would be off but not that off. Wow! I guess that's the last time I talk about Alexa!
So I am curios. If you want to do some invistigation and see how much traffic your competitor is getting, how would you go about comparing sites? Obvisouly you can't ask for their server logs. Do you feel the same way about compete.com?
Oh and by the way, you have a typo in your housepanic link.
Brad... at least I let you know I was stealing it. :) I've made a couple of changes, but I love it!
TechCrunch also wrote about Alexa today. They were even more harsh: http://www.techcrunch.com/2007/08/13/alexa-says-youtube-is-now-bigger-than-google-theyre-wrong/
Brad, I've never looked at compete.com before, but your question is an excellent one. TechCrunch also mentioned them, and at least thought they were more accurate on the Google/YouTube issue. I don't spend a great deal of time looking at my competition's traffic, so I've not really given it much thought. But I'm going to now!
And thanks for pointing out my http error. I appreciate it.
Wait a second. I am not even sure if this is an accurate comparison. I don't think Alexa even showed stats for HousingPanic because it's housed on a sub-domain. In addition, you are comparing a domain to a subfolder of a domain. This probably also throws off Alexa. So the most accurate way to compare would be apples to apples. bloodhoundrealty.com against rsspieces.com I ran those domains and the data is much closer. Now i know you can't really compare domains like this because now you are also including all the pages outside of bloodhound's blog. But I have always used Alexa to compare domain against domain. Shame on you for trying to "work the numbers" to make me look bad! (kidding)
One more thing. I ran the numbers in compete.com and the info looks much more accurate. Remember, am not using this to get a feel for unique visitors. I am only looking at this from a "who get's more traffic" standpoint.
Brad... I tried to act like any average human would, just typing in the url's of the sites on Alexa. I understand the subdomain issue and agree with you, but it only proves the point. You can't simply go type in the URL at Alexa and understand anything about traffic comparisons.
I also just did the same on Compete.com and you're 100% right. They are much more accurate. In that picture, the feel is more accurate. So, from a pure comparison standpoint... I'm going to go test Compete.com out a bit more. Thank you for pointing it out!!
HAHA. Well unfortunately it's a little later over here on the east coast so I am just going to leave and bask in your quote of "you're 100% right". :-) But seriously. Thanks for bringing up this subject. It's important that the AR community understands metrics and knows the good and the bad side of the tools available.
Good night Tim oh, I mean Jeff. :-)
Brad... you're cracking me up. Thanks! I needed it.
Duane... metrics can be tricky, as you can see from the comment stream. :)
As with any set of statistics, caution is in order! Numbers can lie...just as the NAR (haha!).
James... combined with the Google/YouTube comparison, I think we're pretty safe in saying Alexa is a bit off.
Jason... ouch.
Its a shame Brad went to bed...guess it will pick up in the morning. I have always wondered about Alexa. I am going to check out Compete.com too. I like to know what the competitors are doing. Like the saying goes...keep your friends close and your enemies closer.
Informative as always.
Alexa, although not perfect, is a tool that can be used responsibly. What other ranking sites are better? Google and Yahoo Toolbars? Others?
Thanks Jeff!
Rob
Interesting analysis, Jeff. I haven't paid much attention to Alexa as I didn't feel it was worth my time, given what I had read. Guess this validates that impression.
Jeff
that Housing Panic blog is a hoot
man does he go off on people
He took a quote from one of my posts and tried to take me to task - tried :)
Jeff, as usual another terrific topic and post. I wish I had one-tenth your talent.
BTW, just like credits in a movie, sometimes more information is available in the comments. Great discussion here
Jeff,
Great point, Alexa isn't the best method to use when ranking websites. If you navigate to their website here and read the first line in the explanation of their traffic rankings it states this:
"Alexa computes traffic rankings by analyzing the Web usage of millions of Alexa Toolbar users."
The data is skewed by the fact that they only get information from people with the toolbar, what about everyone else that uses the Internet do they not count? Alexa should only be taken with a grain of salt and used in conjunction with other website comparing tools. Great observations.
I'm suprised that no one has mentioned the key problem with Alexa and measurement: It only counts windows users with the Alexa toolbar installed - and the toobar is only in English.
Alexa is the rare exception rather than the rule for international traffic and can't even be used on non-windows operating systems. I'd think that those two factors skew Alexa numbers sufficiently to make the numbers... well...useless.
Someone once disputed our traffic numbers based on alexa numbers. Once they understood the shortfalls of Alexa, they immediately apologized.
Jeff,
I'm so far behind in the on line world! I don't really have a handle on any of this yet. I haven't used any of these tools and wouldn't know where to begin! I can certainly see the benefit but for someone like me who is new to all of this -the idea of organizing the data so that it becomes usable for any given marketing strategy can seem overwhelming and time consuming!! I wish I had more time in my day!! Its frustrating for me being a former EE (hardware) and not understanding this online world better. I feel like I was sleeping the past five years while I was busy being a mommy and technology took over!!! I'm now addicted to reading your blog to find out more information about cool technology that I wish I could learn more about!!!
Thanks Jeff,
What tool do you use for screen capture? Back when I was working/designing, we used a free tool called Snagit. I went to their site and they are not free anymore! Do you know of any free screen capture tools out there?
Jeanette... this works great on the MAC... it's not what I use, but what I use is not free. But I have tested it and it works very well.
http://www.jingproject.com/
You know what would be really fun - Have everyone at AR install an Alexa bar for a week, and see what happens to the traffic here. I bet we could get it on the movers and shakers list ;-)
There really is only one good use for Alexa in my opinion and that is to sell advertising to uneducated marketing people who think it is important - My Alexa rank in the United States is 1,957 in the united states and around 10k in the world, anyone want to buy some way overpriced advertising?
Thanks Jeff,
I wish I was on a MAC....I'm on a PC..but I'm savin for my MAC one day!!! (Incidentally that is my oldest son's name....lol)
Thanks so much for posting this. I have never heard of this, and I consider myselfd fairly savvy on internet sites. I 'll check it out...I just might not download it.
Morgan... care to share what your real numbers are? :)
Jeanette... :)
Bob & Caroline, Leigh, Al, John... thanks.
Roberta... I love it!
I have said before"Alexa sucks" but I still liked looking at it. In fact I am addicted to looking at it. You may have ruined it for me now for all time.
So why does Alexa show RSS Pieces having so much traffic?
That's so funny that Sandy Bauman's daughter is Alexa. If you search ActiveRain blogs for "Alexa sucks" it pulls up three or four of mine and one of Sandi's posts. I could tell right away it was not about Alexa.com and I am sure Sandi did not say anything like "Alexa sucks"that... I will have to go read it though now.
Chris... I need to spend more time on Compete.com, but I was impressed at first pass.
Maureen... it's partially content driven. Because they talk a great deal about SEO, they attract a tech crowd that is more likely to have the Alexa toolbar on their browsers. There are probably other reasons as well, but they will be related to that for sure. Marc has been twittering more since your presentation, by the way. :)
Jeff,
Have you checked out Quantcast.com ?? They seem to be pretty accurate.
That's really strange that Housing Panic ranks around 150,000 on Alexa. From everything I've seen, it's one of the most heavily visited real estate blogs in the country.
I've had my fair share of arguments with people about Alexa in the past... My personal Opinion is that it is fairly accurate for the more popular sites (like in the top 10,000), but once you get any higher, there just isn't enough data to go off of.
Has anyone ever posted a list of the highest trafficked real estate blogs? That would be a fun read. Anyone want to take a shot at who they think the top 5 would be? (I'll bet bubble blogs take up 4 out of 5 of the top spots)... Casey's blog would have been in the running about a month ago.
It's tired in here...
Jeff wrote: "Because they talk a great deal about SEO, they attract a tech crowd that is more likely to have the Alexa toolbar on their browsers." A Dustin Luther comment on RCG was what initially made me find Alexa and most of my comparison has been ActiveRain vs. RealTown Blogs but I always thought Seattle every body and his brother must have the Alexa Toolbar, not necessarily us out in Ohio, Illinois, Indiana, etc... RealTown bloggers are not geeks, they are not going to hang out with Geeks in many cases. RealTown is set in S. California, some tech crowd but most RT blogs are not talking about SEO or technology they're talking about Ohio, Illinois, Florida residential issues.
You may have finally broken me of my Alexa addiction... thanks
Justin... I think you just volunteered. :)
Maureen... give Brad's suggestion a try, compete.com. See how that compares the two. And, you're certainly right about the Seattle connection.
I am not sure that Compete.com is any different or will be in the long run will it? It shows a chart for a comparison of the two that is very similar to the Alexa chart, it knows ActiveRain, is not very familiar with Realtown and so it asks me to put their toolbar on....
Permalink to a comparison between AR and RealTownBlogs.com on compete.com Oh my goodness there are some lovely graphs there though. I also tried RealTown.com (as opposed to just the blogs because I thought maybe with e-PROTalk, RealTalk etc. it would measure higher than just the blogs.
"We have little data for realtownblogs.com, so these are rough estimates. With more data, we can cover more sites. Try the Compete browser plug-in to participate."
Thanks but no thanks. I never put the Alexa Toolbar on either.
It really goes to the heart of how much technology needs to grow before we put all of our eggs in the internet basket. It seems to me that the internet should be pretty easy to measure, but it isn't, it makes no sense. I don't think there is enough push for accuracy, it's all about visitors and the percieved cash value of the site to investors. It really is a shame...
Benn realtorgenius.com
Excuse me if I missed it in your blog but how did you get the unique visitor information?
LaShon.... yes, you may have missed it.
"I went to the real estate segment of topblogsites at Blogflux and pulled three sites from their rankings. I used TopBlogSites because in order to be on their rankings, you must insert code that sends messages to their data collection servers." That is where I pulled the unique visitors from.
Thanks for the quick response, Jeff. The light bulb just went on. :)
Jeff,
Great post. I thought "from SEO people that I have spoke with" the Alexa Rank was the best source of the real rank of a site. I like how you went about with your experiment to test that theory. I think the best results are leads. If you get a lot of them from your web site and blog you're doing the right thing.
Mike Lewis
I've always had an issue with Alexa and how everyone puts so much weight on their numbers. It's good to see this study.
Thanks Jeff!
Jeff,
You are consistently a wealth of information. I have been out of the rain for a few weeks and I know I can always catch up by reading your blogs!
Martin... my pleasure.
Tina... thank you!
Hello i am Mr james the CEO of jamesloaninvestment,i give out loan to an individual who i financially broke who ever that need loan should not hesitate to contact james_loaninvestment@live.com