Google Insights recently rolled out a new feature: 12 month search forecasts. The forecast comes from a relatively simple decomposition of the search volume into trend, seasonal and residual components. The model’s out-of-sample performance is tested on the most recent 12 month period; if that prediction proves accurate, then the model is accepted. Here’s what it looks like when searching for “Google” (blue), “summer” (yellow), and “weather” (red):
The “Google” volume shows a clear macro trend with little seasonal impact; “summer” is of course the opposite. “Weather” proved too unpredictable and no forecast was generated.
A complete description of the methodology is available in this paper by Google (pdf link).
Click here for a live view of these trends.
Hi All,
Todays graph indicates that searches for two seemingly unrelated words, “cliff” and “spark”, follow the same pattern. Now, why is this the case? I believe that the answer is related to the school schedule in the United States. As the graph clearly indicates, searches for “cliff” and “spark” are at their lowest levels during the summer and also decrease significantly around Thanksgiving and Christmas. The declines in the searches for the terms coincide with the time periods that students have time off and are unlikely to do work.
Cliffsnotes and Sparknotes are popular literature and study guides that allow students to write book reports without actually reading any books. It is the searches for these literature guides that is driving the pattern seen in the graph. The increase in the spread between “cliff” and “spark” suggests that Sparknotes have increased in popularity relative to Cliffsnotes. The high popularity of both of these terms in certain areas might suggest that students in those locations are more prone to taking shortcuts when writing papers…
Regards,
R
(Sparknotes was acquired by Barnes & Noble in 2001 – J)
A live view of this trend is available here.
Today’s graph indicates that hunting peaks during the late fall and early winter, while fishing peaks during the summer. ”Fishing” is the more-searched term throughout the year and “hunting” is only more frequently searched in late November/early December. Furthermore, deer is the most commonly hunted animal and “deer” follows the same cyclical pattern as “hunting”.
In most states, hunting for deer is highly regulated. For example, in New Hampshire deer can be hunted by bow and arrow between September 15 and December 15th; they may be hunted with firearms only between November 14th and December 8th. Other states have similar rules. The graph suggests that hunters prefer not to be challenged as searches for “hunting” peak during the time horizon in which individuals are allowed to use firearms.
Lastly, searches for “hunting” were seasonally high in February 2006 due to a little accident involving Dick Cheney, a Bush-Cheney campaign contributor, and a very sneaky quail.
-R
A live view of this trend is available here.
All this talk of Microsoft and Apple’s relative values sent me quickly to Google, where I pulled up the following chart:
The blue line is searches for “Microsoft;” the red is for “Apple.” I’ve filtered it by “Computers & Electronics” to limit the impact of a certain fruit. The blue spike around April first is searches related to the conficker virus, which was supposed to activate on that day. Adding conficker to this graph demonstrates the connection.
Unfortunately, the graph neither confirms nor denys the value trends, but it does exhibit a more interesting behavior – weekly seasonalities! Searches for “Microsoft” drop off on weekends; searches for “Apple” tick up at that time. Could this be evidence for how people perceive the two companies – a work/leisure dichotomy?
Although it appears that the pattern in Apple searches was disrupted – though overall volume was stable – by the iPhone announcement in mid-March. Also note that the search volume was not affected when the two companies reported earnings – not surprisingly, people looking for those numbers tended to search for MSFT and AAPL rather than the firms’ actual names.
I must confess the trend was even more dramatic before I filtered for just computer-related results; apparantly people search much more for red fruit on weekends than during the week!
Click here for a live view of this trend.
Having always been a fan of penguins, I was surprised when I first typed “penguin” into Google Trends. The birds have had a recent surge in popularity, beginning in late 2006. Why is this is the case?
First, I thought that it must have had something to do with the Disney movie Happy Feet. The coinciding of the movie’s release with the upswing in popularity appeared to confirm my suspicions. However, after looking at Google Trends charts for “penguin” and “Happy Feet” in different geographies (as Happy Feet was released earlier in the United States than in the rest of the world), I became unconvinced that there was a relationship between the terms. Another popular penguin-themed movie, March of the Penguins, also failed to explain the surge of interest.
It then occurred to me that Club Penguin, an online interactive game for young children, was released in late 2005 and that the site has become increasingly popular. In fact, the chart below demonstrates that it has become such a success that searches for “Club Penguin” completely explain the rising online popularity of the word “penguin” alone! As a point of interest, Disney acquired Club Penguin in August 2007 for $700 million in cash and future considerations.
In sum, while naively it looks like internet users have become captivated by black and white birds, in fact their children are the ones who are entranced by a website having nothing to do with the Antarctic creatures.
-R
Click here for a live view of this trend.
Update: Thanks to the power of Google Insights, we can provide an even more informative graph (though I wish it had weekly data rather than monthly). The red line represents the search volume for “penguin”; the yellow line for “club penguin”; and the blue line is the difference between the two. In other words, the blue line represents the relatively constant search volume for information on flightless birds (note the seasonality as it peaks each December and ebbs in the summer), after stripping away searches for a website for children:
A live view of this graph is available here.
The title of this post was almost “Tuesday trends are turning into a penguin.” Bonus points for knowing the source. -J
Based on the reaction to Monday’s Google Trends post, The Don (you may also know him as R) is going to help me with a new regular section analyzing interesting trends and insights that can be found using Google’s public search data. We’re going to leg into trend analysis with something relatively simple: extracting seasonality and trend components.
Today’s theme is “uses of fire/burning things.” As you can see, the use of fire/burning things is pretty consistent. However, the target differs by season as people use fire/burn things to make bbq’s in the summer and use fire/burn things to generate warmth and comfort during the holidays. Think of this quick analysis as the starter log for the bonfire of trends to come…
Click here for a live view of this trend.
