Bad Google Trends Data Analysis In 2016 Internet Trends Report?
Post date: Jun 14, 2016 7:33:47 AM
In influential Silicon Valley venture capitalist & analyst Mary Meeker’s annual Internet Trends report for 2016, she states on Page 122 that “Google Trends imply queries associated with voice-related commands have risen >35x since 2008 after launch of iPhone & Google Voice Search.” View the data here: https://www.google.com/trends/explore#q=navigate%20home%2C%20call%20mom%2C%20call%20dad
One cannot infer this from the Google Trends data she shows below:
In the 2016 Google Food Trends Report the data showing a rise on the Trends scale (December 2015 to January 2016) from 57 to 100 (the highest point on the chart - the exact numbers seem to change in Trends as time passes) is interpreted in percentage terms as follows by Google themselves on page 42: "While Mug Cake saw relatively steady growth throughout 2015, it grew +82% from December to January 2016".
In guidance for Trends data, Google states, the "numbers represent search interest relative to the highest point on the chart. If, at most 10% of searches for the given region and time frame were for e.g. 'pizza' we'd consider this 100 (the highest point on the chart). This doesn't convey absolute search volume." They also state, "each data point is divided by the total searches of the geography and time range it represents, to compare relative popularity."
"Google Trends counts aggregate 'searches' ... and not the people who perform them, which a priori does not tell whether a spike in the relative proliferation of a search term is due to a few power users ... Some things are known about the distribution of search frequency among the user population: 70% of searches are produced by 20% of users, and 92% of searches are produced by 50% of users. So 50% of the least active users account for just 8% of the searches ..." - http://wol.iza.org/articles/google-search-activity-data-and-breaking-trends.pdf
A line trending downward or upward in Google Trends means that a search term's RELATIVE popularity is simply decreasing or increasing. As Google say on https://support.google.com/trends/answer/4355164?hl=en-GB&rd=1, “The numbers that appear show total searches for a term relative to the total number of searches done on Google over time. A line trending downward means that a search term's relative popularity is decreasing. But that doesn’t necessarily mean the total number of searches for that term is decreasing. It just means its popularity is decreasing compared to other searches.”
Various studies have shown the difficulty of assessing Google Trends data e.g. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4814066/ & http://link.springer.com/article/10.1007%2Fs10531-013-0552-y.
See my related articles using Google Keyword Planner data on Google Now voice search:
http://www.seolondonsurrey.co.uk/blog/is-google-glass-a-worldwide-failure
http://www.seolondonsurrey.co.uk/blog/what-do-people-search-for-with-their-voice-using-google-now