[charlesarthur]

"In other words, he's saying that half of all users don't do ANY searches at all, whereas all that he can say is that on some days, some users don't do any Google searches on their mobile device."

From the article: "who does what searches isn’t fixed; so someone who did zero searches yesterday might do 10 tomorrow. But equally, the 10-searcher yesterday does none or one or four today. And so on."

So yes. But on any given day, half of people don't do a mobile search. And the numbers from Google suggest that the average across a 30-day period is less than 30. If you think there should be a different model for how many people do how many searches than a Pareto distribution, please offer it.

[chris_wot]

I probably need to re read your article, but where are you getting the assumption that 50% of people aren't doing searches on any given day?

How do you know the spread of data isn't that there are 20% of users who don't use their mobile device for the Internet for 10 days in a row, but then do a lot of searches in a short space of time?

Maybe I'm missing something here, but by taking an average over 30 days, you seem to assume that users do similar numbers of searches every day.