Sunday, November 25, 2012

Converting between Alexa rank and number of site visitors

I think there is a weirdly linear (in a log-log sense) relationship between the visitor count (number of monthly visitors) and the global Alexa rank of a website.
This is called a power law.
It is easy enough to remember - a little like Moore's law, or Ohm's law :)
Relationship between Alexa rank and monthly visitors
It implies that to improve your ranking by a factor of 10, you need 10 times the visitors.
To halve your ranking, double the visitor count.  
If you dropped to 1/3 of your usual site visitors, your new ranking will be old ranking x three.
And so on.

Here's the data I used:

Site %Traffic Reach Est. Monthly Visitors* Alexa rank
yahoo  20 1000000000 4
craigslist  1.5 75000000 42
meetup 0.2 10000000 465
nairaland 0.08 4000000 1385
jobberman 0.02 1000000 4653
cp-africa 0.004 200000 44206
wemabank  0.00028 14000 557445

*The monthly visitors numbers are estimated by assuming jobberman has 1million visitors per month (I think that corresponds to a ranking around 5k) and that the traffic reach percentage (data given by alexa for each site) is simply number of visitors for this site / a fixed number corresponding to all traffic.  That is, take the traffic to be proportional to the traffic reach percentage.


Thomas Carroll said...


I represent a domain name consultancy representing a variety of Fortune 500 brands. I'm very curious to learn more about your methodology as well as whether you have had further success/experience with this topic of converting Alexa rank into visitor traffic.

Do you have some time to discuss further over the phone?

Thanks very much in advance


Tom Carroll

Manager of Reclaim and Research
FairWinds Partners

broda said...

Well defined relationship between visitor and alexa rank of a website.Unique and informative post. Thanks. Alexa Traffic

t said...

Hallo broda, thanks for your kind comments. If you ever want to comment on the Alexa Traffic phenomenon, here, well, sure, completely welcome to group blog here.

t said...

We would expect the graph to shift down and to the right over the years as the internet traffic in the world - that is, the constant (or near constant) product of Alexa traffic rank and monthly visitor count - continues to increase.
If that number was around 6 billion (well, 4 - 9 bn) in late 2012, by now it might be around 10bn say, assuming the web usage in the world has increased by around 10-100%.
* Web usage in this case is an unknown complex of pageviews and site visitor counts (Ref. definition below)

From the page, we have "Our global traffic rank is a measure of how a website is doing relative to all other sites on the web over the past 3 months. The rank is calculated using a proprietary methodology that combines a site's estimated average of daily unique visitors and its estimated number of pageviews over the past 3 months."

It would be interesting to plot a new graph and using the new data, find the new product.

t said...

The "constant product" is now (as of late 2016) about thirteen billion (just over double what it was four years ago)

I also posted here on the math blog

t said...

The "constant product" is now (July 2019) about thirty-six billion.
The data I used for this is for my five blogs, with traffic typically around 1k to 10k monthly pageviews according to my google/blogger "analytics" and the alexa traffic numbers.
1. pageviews are not the same as "visits", are usually greater than visits, which may mean that strictly speaking, the "constant product" is not 36b but something smaller depending on pageviews per visit.
2. I didn't gather data for the full spectrum of popular and less popular sites, I only used less popular sites and assumed/extrapolated that there is still this power law and constant product. Ehn. However, it was a very constant product in the small range I tried, maybe 35b, 36b, 37.5b so, very nearly constant.
You're welcome to comment, add data, update...

Previously on UpNaira