The world’s top websites Vs the 80:20 Principle

I have been doing a little bit of research lately. Readers of this blog may already be aware that I am a big believer of the 80:20 principle, officially known as Pareto’s Principle. If Can’t be bothered visiting that link well here is (another) run down: The 80:20 principle states that roughly 80% of the effect is based on 20% of the cause.

I occasionally think about every day areas of our lives that are effected by this rule. Then over a cup of coffee, or possible late at night after a little bit too much coffee, I go to work crunching numbers and seeing what interesting facts I can gather from it. My last number crunching session while hopped up on caffeine and sweet gooey Macadamia choc-chip cookies found me looking into the Internet world and the websites we so frequently visit.

Delving deep into the statistical data provided to me by Alexa I was focusing on the top 100 websites. I’m sure I would get a more accurate indication using the top 1000 sites but I only have one life time and I don’t plan on spending to doing data entry. Of the top 100 sites I wanted to find figures to explain whether the 80:20 was applicable in the online world. Its one thing to assume it is and another to go out and test it. Firstly Alexa doesn’t give out traffic numbers, only % of total web users that visit a particular site. Not perfect but not too bad either. I decided to use each percentage per website (averaged over 3 months) as points. For example Google receives 29.675% of total Internet users averaged over the last 3 months and I have regarded this as 29.675 points that go to Google. You can check out a screenshot of my list below. (Pictures in my Blog?? Who would have thought!!!)

Data snapshot

Once I entered the top 100 I sorted them into declining order from most points to least. Alexa sites are a bit jumbled up when it comes to this, nothing major but I wanted to keep it as accurate as I possibly could.

So what interesting stats could I make of it? Well first I put the 80:20 principle to the test. Do the top 20 websites make up 80% of the combined traffic of the top 100 sites? I understand that using percentages makes this a little bit difficult but remember we are using the same % variable each time (% of total Internet users). Here is a small pie graph of my findings summary (another picture!).

% breakdown graph

According to my findings the top 20 websites make up 72% of the total traffic points that comprise of the top 100 websites, a ratio of 72:28. I understand that this isn’t exactly 80:20 but we need to remember that it’s a relatively small sample (100 websites). To give you an idea if you pick the top 10 sites the breakdown is 37:62. The top 20 is 51:48. Top 50 is 64:35. So you can see the pattern. If I had the time (or the spider programming skills) and worked this out for the top 1000, that pattern would most likely to have continued to get much closer to the 80:20 ratio. 72:28 is quite telling nevertheless.

Other Interesting facts:

• The top 2 sites, Yahoo and Google, actually make up 20% of all websites in the top 100 list with a total of 20 websites between them
• Google on its own makes up 20% of traffic from this whole list.
• The final 50 on this list only make up 16.1% of the total traffic

I’m relatively pleased that this turned out quite conclusive and I have no doubt that if I was to use a larger sample then that 80:20 would be more definite but the stats show great support of the 80:20 principle.