|
Lots
of people are interested in knowing approximate figures for the
traffic related to an Alexa traffic rank of 100, 1000, 10000, 100000
etc. We've got enough webmasters here to see how good the
correlation is. For any site that you know the data, give the
following information:
a)
page views per month
b) Alexa traffic rank
Don't
name the site. If we can get 50-100 responses across a wide range of
traffic rank then we should be able to draw a scatter plot and draw
a reasonable line of best fit.
I'll
start:
Site
1
a) 12,400
b) 340,000
Site
2
a) 3,000,000
b) 23,900
Next...
:)
It
has been done before.
Try this threat http://www.webmasterworld.com/forum39/893.htm
link in post #14
Also you may try a sitesearch on "alexa ranking". Quite
many discussions about it in the past few month.
From
that thread, there's a link in post 29 to the latest effort:
www.marketing-strategy.info/alexachart.html
but
this could be improved if the chart was on a log scale like the
original sillyjokes one. To my eye, the line of best fit doesn't
seem to fit the data very well.
I
am more than willing to participate again here are my results:
a)
38,240 <- this is the 3 months average.
b) 71,408
HTH,
-gs
Its
obviously not very accurate
a)190'000
b)110'000
a.)
2.000.000 pageviews per month
b.) 11.534 (three month average traffic rank)
Why
do I always have to be an outlier?
Question out of the air. Has anyone noticed a difference between
dynamic and static websites in the alexa rankings?
Uk
sites are always down on US sites for whatever reason
63,000
page views
Alexa rank 625,00
I
have a lot of different sites that are consistant within those sort
of figures , but do not appear to produce anything like the sort of
rankings claimed on WebmasterWorld forums
A
UK site with over twice the pageviews of the above(ie around
125,000) climbs to the dizzy heights of 450,000. While a more
international site with the same page views as this one gets to
225,000 on Alexa. And a third with again the same sort of page views
is around 400,000
You
can probably appreciate why I have little regard for Alexa rankings
:(
<I
do not have an Alexa toolbar!>
Cornwall
- that's probably precisely because you don't have the Alexa
toolbar. One regular visitor with an Alexa toolbar will push a site
to somewhere around the 150k to 400k mark. For mine it pushed it to
250-300k. If your traffic rank is less than that then you probably
don't have any Alexa toolbar user regularly visiting your site.
Hence comparing sites with traffic rank of greater than 100k is not
really worth doing.
However,
on another site that I'm involved with, going from 300 to 20,000
pages listed in Google (solved a dynamic page crawling problem)
shifted our traffic up from 2000 sessions a day to up to 20000
sessions a day and moved our traffic rank from 140k to about 15k.
We're now getting enough traffic to be hit by a significant number
of Alexa toolbar users. Our traffic rank graph now mirrors our
actual site traffic from day to day pretty well.
I
think it's a useful guide once you've entered into the top 100,000.
>>that's
probably precisely because you don't have the Alexa toolbar
I
know that, so that's why I added that I did not have one ;) It may
also be that non USA users are less likely to use Alexa toolbars,
but I cannot substantiate that.
>>I
think it's a useful guide once you've entered into the top 100,000.
However
if you look at say post #4 here with 38,000 page views getting into
Alexa top 100,000. While I find a site with around 4 times the
number of page views barely scrapes into their top 500,000.
It
is basically that "conventional thinking" of Alexa being
eal at top 100,000 that I am questioning. It appears to me that the
"real" value on Alexa probably comes around top 25,000 to
50,000 rather than at top 100,000
I
came across trafficranking.com recently, which appears to come up
with even more dubious figures than Alexa for web site traffic!
Ooh... Not good!
>>Ooh...
Not good!
Yes,
I didn't think you would like that ;)
I
have a lot of sympathy with your original post, but it seems like
searching for the holy grail - you know it's there but damned if you
can find it.
In
theory it should be possible to calibrate web traffic from the info
floating round.
However
the web seems to be so large, that even large sites have stats
skewed by who has an Alexa bar. If we cannnot "calibrate"
the web from Alexa, then the only other person that could do so is
Google ;)
1.8
Millions Page Views
Alexa Three Month Rank - 42,000
Alexa One Week Rank - 18,246
Alexa
ranked somewhere in the rang of 2,500,000th until about two months
ago. Their rankings aren't useful to anybody who knows what they are
doing but I do link to our raking page on our rate card.
From personal experience, having seen a sudden
10 fold increase in traffic and a 10 fold drop in traffic rank,
there's certainly some value in the figures in tracking performance
over a period of time. One off measurements of individual sites are
of less value, especially when the TR is above say 50,000 and if the
sites are in different industries. However, if considering e.g.
webmasterworld against searchengineforums, you would expect a
similar proportion of users of both sites to have Alexa installed,
therefore the comparison is a valid one. Similarly, we use TR to
compare our performance with other high traffic sites in our
industry, mainly in the same country. Our daily TR is now up to
around 10,000 and the daily trend fits well with our personal
website traffic measurements.
18
million page views per month
9,600 Alexa 3 month
7,500 Alexa 1 week
SAD!
I know I'm a skewer on the scale because my users are very paranoid
about security and would never install a toolbar that explicitly
tracks their usage. I'm lucky I register on Alexa at all... of
course, it is still useful to compare to other sites with the same
topic.
It
all seems a bit random really.
I
look at it as a useful comparative tool. Compare your site with
competitors or look at sites where you might want to exchange links
with and might get some traffic from.
Just
a bit of harmless fun
|
Just
a bit of harmless fun
|
|
For
a lot of people it is not really all that harmless.
I
have had companies tell me things like, I can't have an xml feed
because my alexa rating is not high enough, or they won't pay for a
listing/sponsorship because my competitor has a higher alexa rating.
I figure for every person who mentions my alexa number there are at
least 10 who look at and don't let me know.
So,
very once in awhile I have donwload the alexa toolbar and load about
50 pages. That's all it takes to boost my ratings.
By
the way things, things are getting worse. Falsely inflated alexa
numbers are now spouted on just about every eBay website auction.
Not
too many people who understand the value of the Alexa numbers, give
any credence to those numbers under the top 8-10k sites. Sites
ranking higher than 10k - it's pretty difficult to affect your own
ranking. And the numbers are an aggregate over 3 months - not random
surfs.
The
concept behind this thread won't work, because it doesn't take all
the data into account. Data such as number of page views per person.
I also think they may be looking at referral strings and working
that in as a QA test.
Being
in that 8k range, I can attest that there is little way to affect
your own ranking. I spend all day on my site and my ranking dropped
for the week that I used the toolbar. These random spikes and
valleys I see in the Alexa tracking rarely relate to my actual
metered traffic.
However,
since I have a much better ranking than any closer competitor, of
course I use it on the rate cards.
Brett,
unfortunately
very few people understand just how crappy Alexa's data are, but
they still make business decisons based upon them.
"The
concept behind this thread won't work, because it doesn't take all
the data into account. Data such as number of page views per person.
I also think they may be looking at referral strings and working
that in as a QA test."
Well
of course I agree that it's impossible to get precise figures by
back-calculating in this way, but it would be handy to be able to
get a ballpark figure for "I think my competitor is getting
similar traffic to me, 10 times as much, 100 times as much"
etc. If you're only interested in very general figures like that, as
I am, then these calibration graphs can be useful.
For
one of the sites I work with, our daily TR is now around 10k and 3
month average just catching up. Our competitors are mostly between
10k and 30k with one at about 2k and one at about 500. I want to
know a vague figure for how much more traffic the 2k site and the
500 site have than we do.
For
my other site, all the competitors are between 50k and 400k so
there's not much in the way of a conclusion that you can draw.
You
surely can make false Alexa rankings, just look at thearndale.com,
they sell advertising based on their ranking!
They
have shot up from nothing to a ranking of about 5'000 in 3 months!
Also
while your there read a few reviews.
If
you can be bothered to compare their 'old' domain name,
jsearch.co.uk....makes some interesting reading.
If
they can surely anyone who can be bothered to get an alexa ranking
under 10'000 could!?
OK
- I have been prompted into a Friday afternoon jaunt into profitless
fun and I have updated the chart listed in message 3 in this thread
and hopefully will take all the data in this thread and do my best
to improve the formula further.
Although
Alexa is VERY rough and ready, we are starting to be able to see
heavily skewed results and knock them out of the list. The only
alternative to Alexa that I know of is Hitwise - COOL STUFF if you
have a spare US$10,000 per year to see PART of the data. Somehow,
not all of our clients are willing to spend that kind of dosh. But
then... since nobody else has it, I guess they can charge what they
like. Fortunately they still need me to interpret it.
Dixon.
Hi
Receptional,
Have
you considered plotting the x-axis as a log scale? It would move the
points away from the origin and make the graph much easier to read.
Just a thought...
Dazz, that site just went online 3 months ago.
They used other domains to direct traffic to the site. Seems very
legitimate to me.
John
- A log scale works very well I think - much straighter looking line
and makes it easier to discount the outriders. To be honest, my
colleague did the hard work, but I think he did a log table to
eliminate dodgy readings and then put it back to "normal"
with more valid data.
The
"closest fit" formula is really Excel's "best
effort".
Has
anyone tried to convert Alexa reach ranking into actual internet
reach or use reach multiplied by ave. page-views to get an idea of
how many page-views a site has.
For
a rough estimate- I multiply the reach (per million) of a site by
500- guessing that are 500M internet users. But sometimes this leads
to both under- and over-estimates of a site's actual traffic.
I
understand that this method will never produce exact numbers due to
sample bias and re-directs, but does anyone have any advice about
the underlying idea and the choice of 500.
Thanks...
Well,
our reach here says
1400 today. Taking that times 500 would be 700,000. In reality
we are doing about 1/2 that a day.
While
an interesting idea, I don't think the theory holds water.
The
Alexa data is pretty noisy and probably demonstrates a bias from
site to site due to uneven toolbar usage - it's far from ubiquitous,
so a smallish number of users might have a disproportionate impact.
While
that's particularly true for low-traffic sites, even high traffic
sites can be affected. I'd posit that there's a feedback loop at
WebmasterWorld - as discussions about the Alexa toolbar are posted,
more visitors install it, and the site rises in popularity compared
to sites whose toolbar usage percent has remained constant. (For the
heck of it, I put some "free toolbar" banners in rotation
on another site to see if people would click on them and boost the
site's Alexa popularity. It's not really clear what value that might
create, although at some point a very good ranking might be valuable
at some point. Brett can claim WebmasterWorld is a top 400 site
based on recent data - pretty cool bragging rights, and perhaps some
real market value.)
I
like the idea, though, Namezzz - I'd guess there's probably some
multiplier that could provide an extremely crude estimate of
traffic, emphasis on crude.
Alexa
rankings depends on the installed alexa toolbar user base . This
user group i believe will be normally webmaster types and more tech
savvy geek users...
So
Alexa's traffic estimation for websites catering to this type of
users (eg: WW . Slashdot ) will always be an inflated one...
But
for more consumer oriented , mainstream sites the estimate can
approximate to the real numbers
Our
site is skewed very strongly towards women in the US in their
twenties and thirties who should be your run of the mill internet
users. Using Alexa's reach today for us, you would have to multiply
by about 350-360 to get the number of uniques we should see today.
It
would be nice to see if someone could come up with a ballpark
equation for Alexa, but it would probably require a lot of us to
throw some traffic numbers, alexa numbers, and site types out there.
I
also have a feeling that reach and rankings do not follow a linear
relationship with Alexa, but this could only be figured out by
someone with a stats background and a couple hundred data points
from a wide variety of sites.
Alexa
Guy could always just tell us. Brett, when is he joining the site?
It's unlikely Alexa Guy would know. Absent a randomly selected
sample, you'll never know.
I
saw a site that allowed siteowners to enter their Alexa numbers and
their actual daily uniques in order to build a scatter graph whereby
one could estimate traffic from an Alexa number. Several dozen
people had provided their own numbers, and the graph was shaping up
nicely. There were a few anomalies, of course.
As
I recall an Alexa number of 200,000 meant about 1,000 daily
visitors.
This
was discussed on a Jim World site about two months ago. Anyone know
the site that had that graph?
Hmmm...
I think there is a severe breakdown of any multiplier theory under
certain level of visits, or uniqueness of site.
But
isn't there also a "Page Views per user" available?
So
if the "Reach per million users" is 0.8 (stop laughing),
and the "Page Views per user" is 5.5 then... This it where
I find a problem what reach is...
Is 0.8 a percentage, i.e. 0.8% of a million users that is 8,000 so
the page views is 8,000 x 5.5 = 44,000/three months per million
users on the net? Or is it 0.8 for the whole freakin million? which
would mean that my site gets 4.4 page views every million users?!
ARGH!
I have edited this message like a dozen times...
>This was discussed on a Jim World site
about two months ago<
We also had the same thread here. Tried to find it, but couldn't
localize it. If someone remembers, please post the relevant thread
here. I also want to see if others have entered their data. Thanks
in advance.
>
This user group i believe will be normally webmaster types and more
tech savvy geek users...
I
don't know much about Alexa (so correct me if I've got the wrong
idea):
The No.1 site on Alexa is Yahoo, and Google is No.5 I don't think
webmaster types and more tech savvy users would deliver Yahoo as
No.1. - would they? Yahoo sure isn't my No.1 site.
I
found the site displaying the graph of Alexa Rank versus Average
Daily Visitors. Its at
www.sillyjokes.co.uk/alexa/index.php
It
doesn't look like it's been updated since it first appeared. Maybe
by bringing it back to attention, others will contribute their data
too. ( I hope so.. this is neat.)
Now THERE's a URL that inspires a lot of
credibility! ;)
Tapolyai
The
key to your approach (which I like) is to find out how many million
internet users there are- then you can multiply the reach per
million times the page-views times the number of internet users (in
millions) to get an estimate for actual page-views from Alexa data.
Anyone
else have other thoughts on what this might be...
That's
the site! There are about 70 data points on that graph which is
more, I think, than it had two months ago. It looks pretty accurate
to me. Yes, others should add their info. Note that it isn't updated
in real time.
By
the way, isn't that a clever use of the web?
Yes,
the sillyjokes domain gives pause. But I think it's legit.
------------------------------------------
"I found the site displaying the graph of Alexa Rank versus
Average Daily Visitors. Its at
www.sillyjokes.co.uk/alexa/index.php
It doesn't look like it's been updated since it first appeared.
Maybe by bringing it back to attention, others will contribute their
data too. ( I hope so.. this is neat.)"
I really donīt understand one thing... where is Alexa going to get
that data (traffic)?
It canīt be only from the toolbar. Here in Portugal (almost) noone
use the Alexa toolbar, so how can one of our portals (Sapo.pt) be
#287?
The sillyjokes chart's estimate of my unique visitors (based on the
3-month Alexa ranking) is almost exactly the daily average of unique
IP addresses that I had in October, November, and December (as
reported by FastStats).
where
is Alexa going to get that data (traffic)?
It
certainly not perfect but from the bit I've read at WebmasterWorld
the Alexa toolbar is possibly the best indicator of site popularity
on the web because they have a couple of million toolbar users
(anyone know the exact number). That is better than any of the other
official sources of web use.
But how can Alexa get stats from countries
where their toolbar isnīt used or known? It surely canīt be from
the toolbar...
It
is all from the Toolbar. It is much more global than you might
think, despite the language barrier. I found 19 sites from .pt in
the Top 10,000 sites:
Site
Global Rank
1. sapo.pt 293
2. clix.pt 808
3. iol.pt 818
4. terravista.pt 1473
5. abola.pt 3016
6. publico.pt 3141
7. netcabo.pt 3192
8. cidadebcp.pt 3380
9. record.pt 3759
10. aeiou.pt 4011
11. mail.pt 4112
12. portugalmail.pt 4143
13. mytmn.pt 5008
14. ojogo.pt 5765
15. expresso.pt 6635
16. telepac.pt 8565
17. bes.pt 8883
18. vizzavi.pt 9212
19. oninet.pt 9611
Heck,
you could even take a smaller country, with a bigger language
barrier, like Hungary, and find a bunch of sites in the Top 10K:
1.
freemail.hu 1423
2. origo.hu 2027
3. freeweb.hu 4226
4. index.hu 4468
5. lap.hu 5151
6. vnet.hu 8013
--Dunc
Yes I know that they are in the top, I just
donīt know how they get the data to rank it. And thatīs not
because of the language (many people understand english here in
Portugal, we use subtitles in films and start to learn english in
the 5th grade, with 10 years old), but because of the visibility of
Alexa here. Noone knows it here!
Believe me! Only some Webmasters know it, and thatīs a very small %
here.
Where in Portugal could people know about Alexa? Itīs not
advertised anywhere.
My site is #74,349 with #57,000 3 month avg, but I donīt really
understand how they rank this...
I donīt really believe in the toolbar theory for most of the
countries. Maybe they know something through ISPīs...
I have access to some of the traffic of these
portuguese sites, and that rank doesnīt seem well too.
Oninet.pt (aprox 8,000,000 pageviews) has almost the same as Mail.pt
(8,250,000 pageviews), but Mail.pt is rank #4112 and Oninet.pt is
#9611
That doesnīt seem very right...
Check our Ad Agency data:
http://www.hi-media.pt/tabela.htm
I have access to some of the traffic of these
portuguese sites, and that rank doesnīt seem well too.
Oninet.pt (aprox 8,000,000 pageviews) has almost the same as Mail.pt
(8,250,000 pageviews), but Mail.pt is rank #4112 and Oninet.pt is
#9611
That doesnīt seem very right...
Check our Ad Agency data:
http://www.hi-media.pt/tabela.htm
It is interesting to note the the Reach Rank
for the two sites are very close... 7989 vs. 9530. The big
difference is in page views per user where webmail.pt has 12 and
oninet.pt has only 3.4. Apparently the huge difference in page views
is what sets the rankings apart.
Hope
this data was useful to better understand those numbers...
Oninet.pt is a portal, and Mail.pt is a free webmail provider.
Cyas,
Nuno
Great
thread!
It
is perfectly understandable that a web mail account could have
higher page views per day than a portal page - because users may be
checking their mail many times per day.
But
it is also conceivable that it would be the other way around.
Consider the Yahoo home page which is likely to be loaded many times
per day -- users who have it set to be their home page load it every
time they close and re open their web browser - this could occur 50
to 100 times per day per person!
Couple
of more thoughts...
Epistemology
-- the branch of philosophy which investigates the limits of
knowing. IOW, it asks the question, how do you know that you know
what you know? This concept is at the root of all web tracking. All
web stat, traffic and tracking questions ultimately boil down to
questions of epsitemology.
Example
of epistemology: you are sitting in a chair in a room as you read
this. When you get up from the chair and turn your back on it, how
do you know that the chair is still there? Perhaps it disappears
when you are not looking at it and reappears when you are looking at
it.
We
take it on faith that the chair is there all the time, because our
experience tells us that this is likely to be so -- in other words
we assume it to be true, but we don't in fact actually know it to be
true. In fact, there is no way to prove that the chair is there when
no one is looking at it or touching it.
So
it is with web traffic counts. You only know about it because you
can "see" it in a report or log or some other
manifestation.
We
make many assumptions in
doing so - we assume that the server is counting every hit, we
assume that we are capturing the right information in the logs to
distinguish between hits, page views and unique visitors. We assume
that Analog, WebTrends, HitBox, or other tool is reading the logs
correctly and not introducing errors of its own when analyzing the
logs.
We
assume that the reporting engine or graphing tool or OLAP tool is
processing and displaying the data accurately and not introducing
errors of its own - but all of these are basically proprietary black
boxes and we will never have any way of knowing.
And
perhaps the biggest assumption of all - that it is necessary to
measure every page request coming into the server, and count the
behavior of every visitor, as opposed to a representative or random
sample of requests and visitors.
And
all that is just considering our own site. When we want to compare
one site to another, we are really comparing our assumptions to
their assumptions.
From
a logical analysis of assumptions, we arrive at skepticism- a
questioning or disbelief that something is true, especially those
things that we assume are true but haven't or cannot prove.
And
we have seen posts from many here who are skeptical, for instance,
of the idea that Alexa is a valid measurement of absolute web
visitor counts, and therefore skeptical of the accuracy of its
rankings.
At
the heart of the skepticism is a question - how does Alexa know what
it knows? Since we do not know the answer to this question, we are
skeptical of Alexa's output.
Brett
expressed skepticim about Nielsen ratings, questioning the accuracy
of its sample sizes and questioned their ability to accurately
represent all web users.
But
the fact is, as skeptical as we may be, when you add up all of these
assumptions, there is no way you will ever have the time or
resources to systematically track down and verify each one of them,
not to mention resolving all of the issues that you find.
On
the other hand, by learning to use the tracking tools and reports
that are available to us, through experience, we begin to feel
comfortable with the assumptions that are built-in to them, even
without knowing what all of those assumptions are. We accept the
truth of the report even though the truth is a distortion of
reality. Eventually, just as we trust that the chair is always
there, we can trust the reports. Like riding a bike with crooked
handle bars - we can learn to steer it.
But
we don't necessarily trust the reports in an absolute sense. Rather
we trust them in a relative sense by observing patterns - patterns
of change over time, and relative rankings.
The
entire subject of search engines and rankings of any kind is fraught
with these issues. The reality is we will probably never know
certain important methodologies, for example, search results ranking
methodologies.
We
should at some point, perhaps a few years down the road, be able to
know what the popularity ranking methodologies are, and be able to
compare, for instance, Nielsen's methods to Alexa's, but it will
probably take some official independent research group or academic
study to do it.
This
will only be fair, because search engine methodologies were dreamed
up by Phd's. Now we need some new Phd's to study what these other
Phd's have done and tell us how they how they did it.
The
Alexa chart at the sillyjokes site looks pretty accurate for the
stats of some small and large sites that I am familiar with. It
correlates ranking with traffic to within 10-15% of the
"actual" traffic counts that I have recieved.
As
for Alexa's rankings, due to the large number of assumptions that we
each make in trusting our own web reports and the huge opportunity
for (Error X Error) that occurs when trying to compare Web Site A's
reports to Web Site B's reports, there are tremendous benefits to
having a universal third party measurement system when it comes to
comparing visitor counts and page views of different web sites - by
measuring all sites by the same criteria, errors will be cancelled
out, a major advantage.
Even
if their methodology is flawed, if they are applying it universally
to all sites they are counting, we would be able to rely on it as an
important relative measuring tool.
I know many people that have changed their
homepage from a portal to a webmail. Itīs more likely to consult it
more often than a search engine.
I agree with all your post, and... Welcome!
FYI
- Alexa used to display the number of visitors as well as rank.
About a year ago [I think] they stopped doing this.
That would be even more difficult to
understand, how they could give an aproximate number of visitors...
but it would be easier for us to compare with our own data.
Welcome too ClayCook!
I
have updated the data from sillyjokes' site (with their help) and
put a revised chart up at
www.marketing-strategy.info/alexachart.html
Hope
Brett doesn't shoot me for putting up the URL - there was a thread
somewhere where we were collecting data, but can't find it now :)
|
