This week’s installment of ‘Jakon Nielsen’s Alertbox’ has the title of “Data Visualization of Web Stats: Logarithmic Charts and the Drooping Tail“.

I got all excited at this post (before reading it) because I thought Jakob would be taking on visualization usability (is that even a concept?) however, Jakob seems to be more interested in challanging the ‘Long Tail’ theory than discussing visualization. Not that Jakob totally disagrees with the Long Tail theory, he just seems to think that it’s difficult to exploit and only a few uniquely positioned companies will be able to exploit it in any realistic way. Ok, enough paraphrasing, here’s what Jakob concludes:

It probably wouldn’t pay for our sample site to take advantage of the opportunity that log analysis revealed. The long tail’s end pays for aggregators who get their products from others, but companies who must develop their own are usually better served by staying away from the full tail.

That said, pursuing the tail’s end might be valuable if a site meets one of two conditions: it has a better way than low-value ads to monetize traffic, or it has so many users that the total income would be substantially more than the cost of developing the new functionality.

In any case, you should certainly run through such exploratory ROI scenarios for your own site. To do so, you need correct data analysis and this typically requires more advanced visualizations than you see in most places. It’s here that logarithmic plots deserve a chance — despite their intimidating name.

Ok, back to my point, Jakob’s visualizations aren’t particularly advanced, what is advanced however, or at least more advanced than what we tend to use in our everyday work, is the application of Zipf’s Law and the use of a logarithm to show how a site has a ‘drooping tail’. Which proves that if he wants to exploit the tail his example site needs to perk its tail up a bit.

I guess the key take-away for me is (and it should be for you) make sure you understand the theory (any theory) that you are applying to your site or web analytics and do it correctly. If you don’t you’ll most likely reach the wrong conclusion and that is going to be painful.

The great thing about a theorum, lemma or other model is that if you can measure something in relation to it you’re going to quickly see ‘areas of interest’ - e.g. where emperical evidence does not match the predicted value.