How Meaningful Are Alexa Data?

I can read the Alexa explanations on their site, but, well . . . Alexa wrote them.

How much significance should one attach to Alexa's reports of traffic data, notably: traffic rank, reach rank, and page-views rank?

I ask because, while I'm getting my toes stepped on in the Google dance, Alexa--to my pleasant surprise--shows me steadily and continuously powering up in all of their data (except page-views/user). My 3-month traffic-rank average is #706,164; my one-week average is #544,248; and my rank today is #269,483.

(Meanwhile, the long-time consistent Google #1 SERP for my keywords (I have hit #21 but am currently #38) has an Alexa 3-month traffic rank of #2,503,268, which is down 1,482,691, and they don't even have a weekly or daily report rank.)

So should I look at Alexa and rejoice, or look at Google and weep?

 

Hi,

it's very relative as the usage of the toolbar is still smal.
BUT it can be a good indicator if you compare it to positions and movement of peer or ASP's (alternative supplier pages) and assuming they don't manipulate it.
It becomes more reliable and interesting if you site can pass 100.000. The top 25.000 show a good indication of overall traffic trends.
Don't forget its Amason and the ia_achiver feeds the way back repository which can be embarrassing but also very helpful if you ever have copyright / design / name issues.

I also think it has been covered in another thread and you may want to do a search.

Cheers
M

 

I wouldn't count on it, very few people beside webmasters are using it and if the webmasters are not your target audience then it does not reflect much.
I would more concentrate on the on-site logs to see trends
/BP

 

Hi Owlcroft,

Alexa is a waste of space, unless the demographic is correct. My example (somewhat recent actually) - 3K+ unique visitors per day. No registration on the chart.. i think somewhere around 125,000. Then, all of a sudden, I jump to 30,000.. why ? because webmasters from the US/UK were responding to a link campaign and checking out my site for reciprocation etc. Around 1200 I think the number was over a period of two or three days. The site I'm referring to is an Australian real estate site..

So demographics are everything in the case of Alexa. I think you should (to reiterate what others have said) rely exclusively on logs. If someones trying to sell you traffic on Alexa figures, don't go there Of course, there's software around that's custom designed to crank the figures as well...

On a brighter note - MAN what a house ! I love it ! I'm not a gardener, but you've got yourself a private paradise right there. Library.. fireplace.. cellar.. aspect.. I especially liked the contrast photos, ie inside, outside etc.. very nice

 

First, thank you John for the thoughts about the house. We designed it ourselves, and are most pleased with the result. We got taken to the cleaners by a crooked builder who abandoned the project half-way through, so we had to get someone else to finish it at an overall cost that seriously impacted our retirement, but at least we like the product.

Meanwhile: my problem with site logs, which problem may well be an artifact of ignorance, is that when I look at all the data (using Webalizer), I see no straightforward way to eliminate--or even identify for certain--searchbot visits, which are a large part of the hits count, so as to get to "real" visitors. Any expert ideas there?

Alexa just seemed like a moderately useful barometer of trends in "real"-visitor counts. It also seems useful for making gross (not fine) distinctions between sites on popularity. I do not attach much significance to a claim that this site is #66,000 and that one #68,000, but I do believe that there is a major difference between a site they call #67,000 and one they call #2,500,000.

But I say "seems", which is what I was and am inquiring about. Is Alexa useful even on a gross scale?

 

Good to hear you like your home. I hope I'm able when I retire (a few decades off yet ) - minus the crooked builder of course

It's 1am so I'll be brief - Alexa on a **gross scale** will of course reflect some differences for you - ie theres a diff between 2,5M and say 300,000. But that difference could vary between 40 Unique visitors a day and 1000 (perhaps even more) - it comes down to how many of your visitors are using the toolbar, thus reporting the visit to Alexa.

Next - your log (your only real option).
Webalizer is handy, but again, rather difficult if you want a quick count of uniques. You could do it manually ie grab a list of robot agent strings (see below), and remove all matching agent string entries, remove duplicate IP's and you're left with your uniques. For more on active bots go here; http://www.robotstxt.org/wc/active/html/index.html

The other option, hunt around the net and find yourself a tracking service, web apps/tools. There's plenty of freebies and they're pretty easy to implement. I know this can be a hurdle (especially filtering bots from everything if you're stuck with whatever your provider gives you).

One example is : http://www.reinvigorate.net/
Handy metrics etc, but in order to filter bots out, you need their IP's.

Optionally, you can go for professional reporting software (that digests your logs) such as WebTrends.

Well, hope that helps. Nite

 

This is from someone who doesn't know a whole lot about the tools and gadgets available to pump up ranks. But if I were an advertiser Alexa is the only tool useful to me. (maybe weblogs, however you get those). My traffic grows everyday and I can see that on Alexa. My page rank does nothing but people still come. Advertisers don't care where your banklinks are coming from, only that you have traffic. If I put a backlink on a site, am I doing it to pump up a page rank or to draw more people to my site?

The most impressive stat I see on Digital Point is it's 1455 Alexa traffic rank. If I want to advertise on DP, the 6/10 or # of backlinks doesn't mean anything to me. I want to know how much traffic it gets. My current Alexa 3 month rank is ~131,000 but it was 4,000,000 3 months ago. My daily rank is usually 60K to 100K. During that time my PR has remained 1.

Like I said I may be looking at this from a different perspective. I still pay AdWords $10 a day, but I get ~3000 visits a day. Nothing I have done has helped my PR but if I up my AdWord to $30 a day, I get 3 times more traffic. Of course if you are in a competitve market you can pay a lot more than that. I still get top billing with my nickel a click bid.

Now, who wants to help me with backlinks?

 

I'm going to ramble a bit about several thing that have been brought up in this thread.

1. Pure and exact numbers for anything on the Internet are almost impossible to attain or verify. So what would the removal of the SE bot's visits from Webalizer stats do for you? Assumming the bot's visits are more or less constant what you want to see is an upward trend. Absolute numbers unattainable and unnecessary.

2. To suggest that Alexa numbers might be better than Webalizer is really a stretch. Webalizer is working with the traffic logs that apply to your site. Alexa is working with a very small subset of the traffic to anybodies sites and their measure is a broad generalization at best and may be totally useless at worse.

3. Having said this about Alexa, things may change a bit. The new Amazon A9 search engine uses Google's results and works in conjunction with Alexa. So depending on how many people migrate to this search facility -- and I like it BTW -- then Alexa's results may become more useful and meaningful.

 

I would look at your site stats!
For my site, I've noticed that Alexa numbers dont match the traffic stats I've received. My traffic has not changed, but yet Alexa says that I'm getting FAR LESS traffic! That can't be, and maybe because I've uninstalled the Alexa toolbar may have something to do with it? I dont really know. Now I dont trust Alexa as much as I did before.

 

Alexa can be manipulated if you so want.

Your own statistics tell precisely (more or less) whats going on at your site.

BUT Alexa has its uses. 3-4 competitors show an upwards trend, so why does my/your site not show a similar trend or is it exceeding it......

By looking at sites in the top 50.000 which are either in or closely related to your industry you can see seasonality.

Example travel (Europe): mid first quarter shows a spike. If you ride that spike your good for the season - if you missed it, very hard to catch up.
20% in Q1 translate to 25%+ for the rest of the year.

Thus Alexa used over many sites gives some great indications but thats all there is.

Also if you know your competition average page views can be usefull to determine their overall direction / content acceptance.

Expat

 

alexa rankings have an advantage over the pageranks for they are not a mere segregation in 10 sectors but a complete no. Though the later gives much more information on the site statistics

 

Did you just reply on a 6 year old thread?