Swine flu — there is at least one good thing about it…

We are in the midst of an information storm about an apprehended flu pandemic. Data clearly shows a spike starting 4-5 days ago, in the news as well as in the blogs, where the number of indexed documents referring to flu, of any kind, shows a ten to hundred–fold increase. That is not surprising, would you think. But what is, at least to me, is that Google trends' special flu microsite shows absolutely no sign of increasing activity.

This is odd — I would have bet the house that people are searching [flu] is near record numbers. In fact, I've added [flu] to the list of concepts I track precisely in order to examine the lead-lag in search vs visibility (i.e. do people search for complements of information on what has recently become visible, or does search signal interest which is a precursor of visibility?).

Unfortunately, "search for search" is much more difficult than searching for visibility. Looking at Yahoo! Buzz, I see that swine flu was the top search during the past hour. Looking at Google trends, the current spike is obvious.

So, yes, people do search for complements of information, like crazy. And this is certainly a case of reactive search (i.e. a news report has initiated the process). The next question is why is this surge not showing on the flu microsite? Is it sophisticated enough to distinguish between search terms revealing actual infections (such as [I am sick with the flu]) and queries motivated by curiosity?

One plausible answer is that the flu microsite is updated once a week and that the spike will show up at the next update. If this were the case, then we will have to reconsider the punch line that Google trends provides a two-week lead in the number of flu cases reported by the CDC. What was "white magic" (as in white hat hackers — the use of the "infinite power of IT" to do good) would take a blow as the number of search would just be "experiential coincidence". (i.e. people search for flu just because they anticipate getting sick, just as people would be more likely to search for [sunburns] during the summer ; people search for carribean beaches *before* they go on holidays, and so on). 

So I am eagerly anticipating the microsite updates as I am curious to see how sophisticated this search analysis is. And if it turns out that it is, I will then worry that this information is proprietary. And if it isn't, I will be disappointed by the limited wisdom of the crowds… So it is win-win 🙂
Publicités