Forum Moderators: Robert Charlton & goodroi
a new layer to Trends with Google Trends for Websites, a fun tool that gives you a view of how popular your favorite websites are, including your own! It also compares and ranks site visitation across geographies, and related websites and searches.
I wonder if they're using Google Analytics data for this? It doesn't seem so on first glance.
The Google Analytics data in Trends for Websites comes from the anonymous opt-in data sharing setting in Google Analytics. This setting enables website owners to share their data in an anonymous form.
I came across a couple of my own. Scary accurate. Its also pulling some of the top search phrases and showing those as "Also searched for..." Hey, no fair, foul!
Try checking the trend data for google.com -- "not available" -- now that's double standards BS if I've ever seen it.
share their data in an anonymous form
If that means Google trends shows actual traffic data taken from G Analytics for a specific site, I'd hardly call that anonymous!
This being said, it's a great tool!
For a long time I've wanted some method of checking overall Internet traffic levels, and this seems to fit the bill. Since "google.com" doesn't work, I've searched "Yahoo.com" and it appears that traffic there (especially 30-day trends) correlates pretty closely to my main site.
Now when I see a slight uptick or downtick in traffic, I should be able to tell within a few days if it's an issue with my site or if it's an issue with Internet traffic in general.
Scary accurate.
So I decided to open GA and GT side by side while reviewing various accounts that are being recorded in GT.
Regions information appears to be dead on.
Also visited, let's talk more about this one.
Also searched for appears to be a random sampling from the top 10-20 searched phrases. There is no order but the phrases appearing are in sync with GA.
Now, login to your Google Account to get the "Numbered" graphs. :)
Google Trends kind of makes Compete and the others a lesser tool since we all know that the data Google is showing is likely the most accurate of all of them. Compete.com doesn't have their ga.js on millions of websites!
This should help Internet Marketers out a bit. It gives you a good insight into the data being displayed. Heck, I would imagine this is a link developers dream? Since Google is recording all those Also visited statistics, I may have a good solid list of ten (10) properties to look at for partnerships, whatever they may be. ;)
Certainly it's an annoyance (and suggestive!) that Google properties are excluded (e.g. YouTube).
But of course, the competition can do the same for my sites, and that feels really bad. I've never recommended data sharing for Google Analytics, and I certainly don't now. But the level of data they show even for sites that do not use GA shows that GA data is not the main culprit.
In short, I am not a happy camper to see this kind of data exposed so easily and casually. I'm very tempted to go into a rant.
Clearly, there's a lot of commercially sensitive information available through a tool like this. But I have mixed feelings about whether publishing it is a good or a bad thing. You can already buy comparable data, right?
My initial feeling is that the additional data is 'webmaster tools'-ish - i.e. mostly based on Google.com activity. But if toolbar data is integrated (which it appears to be) then that's pretty significant.
I feel somewhat aggrieved that Google chose not to publish data about their own properties though. IMO it can only be described as hypocrisy.
Slightly off topic: Google could do a lot of good in extending their policy on other people's information to themselves. And why not publish browser stats/javascript usage and so on which would be of great use to site owners? There's clearly no harm to Google in doing so.
I must say that it is nice to see some 'popular' sites being shown as not as 'popular' as their owners might wish us all to think
Note that the traffic numbers are not entirely accurate. As with any of these tools, the more data they have, the more accurate it is. So, the majority of sites (who receive little traffic) won't have accurate data.
But, it does help site owners who are tempted by advertisers misrepresenting their popularity (something I've come across frequently).
And you know I tried Google.com when that link was first posted. How dare them! Where's the Beef? Snicker, snicker, snicker...
Now when I see a slight uptick or downtick in traffic, I should be able to tell within a few days if it's an issue with my site or if it's an issue with Internet traffic in general.
I'd be careful if you're using the data that way. There appears to be an underlying benchmark that is applied to all websites (with varying degrees of accuracy).
In the US, what is the difference in your area on searches on:
<your city name, State spelled out>
and
<your city name, State postal code>
And,
do people search on "Cellphone" or "cell phone."
The problem in using the tool is the question as to whether Google Trends is over or underreporting traffic to other sites by similar percentages.
The Google Analytics data in Trends for Websites comes from the anonymous opt-in data sharing setting in Google Analytics. This setting enables website owners to share their data in an anonymous form.
I'm wondering if opting "IN" to GA temporarily just so you can check your stats, and then opting "out" again, just how long your site will be available for comparison. Anyone tried it?
One really amusing thing I found is that the bottom drops out of the traffic in December for even the biggest sites. I'd noticed this trend in my sites over the years. Proof that many women have no time for their own hobbies during this period. All their time goes to getting ready for Christmas. I find my best earning month is always January. Spending their Christmas money on themselves I suppose.
It seems to me that ONLY the sites which are being added by webmasters in their Google login are showing any sort of data.
Frankly, because of this application of Trends data, I'm relieved that the reported data is as limited as it is. I'd rather see Alexa and Compete fumble around on this and have Google stay completely out of it. It's one of those areas where Google feels like fox, hen-house, and Big Brother all rolled into one. I don't want competitors to be able to pick up on my strategies so easily.
For a widget company we know that maximum traffic will come from blue widget, red widget (and all possible combinations) ... the competition is tough for these keywords as well but now its revealing the secrets ... name of a bad widget company which does not even have the widget word in it (you have reviewed them) and you get maximum traffic and links for this keyword.
I know some of the sites where it was a big secret till now, now it is flashed right on screen. Competition (esp PPC) is more open now.
It's very poor for any site that does not use analytics. It's almost creepy in a "use our product or we will dis you" way. Quantcast is similar in how if you add their pixel, you get higher traffic numbers (and more accurate).
So there is a tradeoff... don't use Analytics and get under-reported which can hurt you with advertisers (someday anyway); versus, use Analytics and have all your prime search terms revealed to the world.
For US traffic, Compete and Quantcast seem still much better tools (since most sites are judged on a level playing field), but for international data Google at least seems better than Alexa (which should just go out of business already it is so lame).
So to sum it up, Google again releases a crappy product whose best feature is that it is less crappy than the crappy competition.
For 2 websites that are in top 100,000 according to Alexa, there is only 1 keyword listed. There's also GROSSLY overestimated international traffic vs. US traffic (percentage wise US traffic should be much higher). I am guessing because of the low install base of a toolbar or whatever spamtool Google uses they have to algorithmically extrapolate international traffic, which is as inaccurate as it gets.
Luckily for us, no keywords revealed. Not yet, anyway :-)
That doesn't seem right.
Looking at "All Regions" I see just almost 30 million, but 6 out of 10 also visted a .jp site?
Hmmm, did they possibly uncover something going on over there at the Wiki?
