So who isn’t in love with Google’s competitive data tools for web marketing? Of all the sources of public web analytics, Google is potentially the most accurate. Why? No one else has the breadth of web data that Google has. Not comScore nor Nielsen. Not Hitwise. Not quantcast, compete nor Alexa. Forget their admittedly impressive press releases. They just don’t collect anywhere near the about of data Google collects. Consider:
- Google tracks many sites directly with Google Analytics
- Google samples many sites through web users who navigate with a Page Rank enabled Google Toolbar which “calls home” to Google in order to display the Page Rank
- Google knows how much traffic websites receive from the world’s number one search engine
- Google knows how much traffic websites send to sites using Google Analytics
- Google provides a DNS service
- Google samples traffic to sites in the Google contextual advertising content network, otherwise known as adSense, in the process of displaying those ads.
Add these data points together and Google is able to build a pretty accurate picture of what is happening on the web. “Free” is also a great selling point, although I wouldn’t recommend making significant business decisions based on the data from just any one supplier. Verify, verify.
One of my favorite Google tools is Google Trends, launched back in 2006. Particularly interesting is the website traffic volume comparison feature, introduced in June 2008. It is available through an easy to miss link at the bottom of the Google Trends home page.
The following graph of traffic to Italian newspapers was captured last summer.
I just updated the graph for my SEO course:
One thing in particular leaped right out at me. Data which went back to May 2007 now only goes back to November 2008 (keyword research trends still go back to 2004). Google’s Ad Planner also showed data back to May 2007, at least in a screen shot I took last November. Yet the renamed Google Doubleclick Ad Planner now only shows traffic back to the end of 2008, which leads me to believe that Google reduced the data available in Google Trends for Websites at some point between last November and now.
Dear Google, historical data has its value. Please consider restoring it!
Oh, while you’re at it. please don’t violate net neutrality by doing evil deals with Verizon and other cellular carriers. Cellular internet, despite health and other concerns, is the ONLY connectivity option in many places besides analog dialup. Big companies should not be able to decide who and what gets priority on our internet. Their commercial concerns are far too often not aligned with user’s best interests.
Similar Posts:
- What Google Knows about Google… and a few other things
- Top Keyword Search Trends in Google, Yahoo!, MSN and Ask
- 10 shadows hanging over Google: does Google have a dark side?
- How do search engines, such as Google, handle JavaScript and CSS?
- Google Zeitgeist has a Sibling, Google Trends
Registration is now open for the next SEO Course (March 22 and 23) and Google Analytics Course (March 14 and 15) in Milan. Don’t miss the opportunity!
Sean Carlos is a web marketing consultant & teacher, assisting companies with their Search (SEO + PPC = SEM), Social Media & Digital Media Measurement strategies. Sean first worked with text indexing in 1990 in a project for the Los Angeles County Museum of Art. Since then he worked for Hewlett-Packard Consulting and later as IT Manager of a real estate website before founding Antezeta in 2006. Sean is an 
I’m fairly certain that Google changed how much historical data they report because of changes in methodology. After major changes in methods, historical data and current data would become incomparable.
Sean’s comment is correct. We updated our traffic estimation models back in Feb, 2010 to more accurate models.
Blog post about the update:
http://doubleclickadvertisers.blogspot.com/2010/02/better-site-traffic-data-in-doubleclick.html
With the traffic model update, we had to recalculate our historical estimates and we could only go back to end of 2008 because of Google’s data logging policies.
@Sean, Wayne, thanks for the feedback.
I’d say that the title is misleading in some way. When I came to read the post, I assumed that less information would be available. However, what you mean is that there is less HISTORICAL data, that is, that the dataset does not go back as much in time.
Based on what Sean and Wayne, state, I would however assume that this is only temporary, and that historical data will grow again as soon as they are available. (provided there is not an additional algorithm change, of course