The writer cautions that we do not know what GoogleTrends is reporting: Bengo initially thought the data came from Google Analytics, the free platform that Google provides webmasters, but added later that this is not the case. GoogleTrends describes itself in this manner:
Trends for Websites combines information from a variety of sources, such as aggregated Google search data, aggregated opt-in anonymous Google Analytics data, opt-in consumer panel data, and other third-party market research. The data is aggregated over millions of users, powered by computer algorithms, and doesn't contain personally identifiable information. Additionally, Google Trends for Websites only shows results for sites that receive a significant amount of traffic, and enforces minimum thresholds for inclusion in the tool.Bengo also adds that several webmasters have reported that what GoogleTrends has reported for their sites is not reflected in their internal data, so there are a number of grains of salt in all of this. It's of note that you can't pull up a GoogleTrends report for Google itself — so if the number of people using the Google search engine dropped, for example, everything might move with the tide.
That said, we can look at the GoogleTrends charts for some other sites and see things that, we might surmise, reflect reality. Regard the most basic chart for Marvel.com:
...which displays clear spikes for the Iron Man movie and the second Hulk movie. My own experience with website traffic tracking is that it's good for the showing the broadest of moves, like this one — and for micro-level information, like who's coming from where on a particular day. But comparisons between months and, most particularly, between sites are subject to a lot of other variables, often having more to do with what's being used to do the measuring. Alexa site rankings below 100,000th place see ridiculously wide swings — and GoogleTrends as well doesn't report on sites where it hasn't got enough data to reliably sift.
That said, what would it mean if the traffic of major webcomics sites were, indeed, softening as a group? Bengo suggests one driving factor might be the rise of the ComicPress theme for Wordpress: "Big host sites are losing out to owner operated sites. WordPress and other tools make it easier to be less reliant on hosts. This doesn't mean big host sites aren't able to grow, but without enough good comics calling them home, their all-important social communities will fracture."
I would also look at the possibility that advertising platforms like Project Wonderful are allowing webcomics creators to distribute their links far more efficiently, and draw upon the larger existing pools of readers, effectively draining off a portion to subsidiary lakes. (I have some direct experience with this: My own webcomic with Chuck Fiala has seen more traffic in a week from the ad system than it has in whole months before. The lake was landlocked; now, there are canals.) You might have a situation where the number of online readers for comics is increasing in aggregate, but where the linkages are making it easier for readers to migrate the sites that most suit their interest. Thus, an increasing fragmentation of audiences, just as we see in television.
The print-comics equivalent would be the direct market distributor catalogs of the 1980s, which put small-press efforts alongside the established comics from Marvel and DC. Exposure to those larger audiences established many new publishers — and the number of copies sold went up in aggregate. (However, in the 1980s case, sales at the majors were not really harmed by the wider selection — and in the early 1990s, while the majors did lose market share, their own sales still went up with the rest of the boom.)
It will be interesting to hear how well the GoogleTrend data holds up as people compare the information with their internal and other external metrics.
Update 3/12/09: And I have heard from those people, and, as Bengo heard, the GoogleTrends numbers are quite a bit lower than what people are seeing internally. Read more here.