Sogou Explorer and Baidu Browser are hidden in Google Analytics

If your website serves the Chinese market you’d want it to work well and render properly for your customers. I was checking the browser stats for a Chinese website and my colleagues were surprised to see that two of the popular Chinese browsers are missing from the list: Sogou Explorer and Baidu browser.

So I installed them to check what’s exactly going on for these two. Let’s start with Sogou Explorer by opening the developer tools console on the Network tab (you can check Preserve log box) and loading our website with custom campaign parameters like ganotes.com/?utm_source=test&utm_medium=test&utm_campaign=test. Search for ‘collect’ which is a part of the GA hit and look at the headers:

Sogou user agent

The user-agent shows a long string: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/49.0.2623.22 Safari/537.36 SE 2.X MetaSr 1.0

When we go to Google Analytics to check how the traffic from this specific source / medium is recorded we see:

Sogou in GA

and the browser version is:

Sogou version in GA

So, in my case my session from Sogou Explorer is recorded as a session from Chrome 49.0.2623.22, that’s why there’s nothing mentioning Sogou in Google Analytics Browser report. Depending on its internal logic, sometimes it will show up as Chrome and sometimes as Internet Explorer.

Now, let’s do the same for Baidu Browser.

Baidu user agent

The user-agent shows a similar string: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/43.0.2357.124 Safari/537.36

When we go to Google Analytics to check how the traffic from this specific source / medium is recorded we see:

Baidu in GA

and the browser version is:

Baidu version in GA

Similar to Sogou Explorer, Baidu browser is recorded in GA as Chrome 43.0.2357.124, and there’s no way to distinguish between real Chrome sessions and these ‘fake’ ones.

How to troubleshoot click tracking with Google Tag Manager

It’s easy to make a mistake when you setup your click tracking in Google Tag Manager or you just may not know how to define the click triggers.

Here’s a way to understand how GTM sees your click elements and end up with a properly working click tracking.

Let’s say we have a button that we want to add click tracking to.

First, go to Variables page in GTM and make sure all Click variables are enabled: Click Element, Click Classes, Click ID, Click Target, Click URL, Click Text.

GTM-clicks-variables

Next we want to create a simple click event for all clicks.

1. Go to Tags – New, give it a name, and select Google Analytics and Universal Analytics as product and tag type.
2. Enter your GA tracking ID or use a variable if you have one. Select Event from the track type drop-down and give it some generic category and action values.
3. Now for your trigger in the Fire On section, select Click, give the trigger a name and select All Elements for All clicks.

GTM-click-trigger

4. Save your tag.

GTM-generic-click

Enter Debug mode and refresh the page of your website which has the button you want to track clicks on. Let’s say I want to track the clicks on the comments balloon icon on the blog. Click the button while pressing Ctrl key so it opens a new tab, and let’s see what the Debug console looks like:

GTM-debug-mode

So we have our click tag firing. Now click gtm.click in the Summary pane on the left and select Variables at the top. Here we can see what data layer variables GTM sees for this element so we can use them to improve our trigger to track clicks on this button only, and not all clicks:

GTM-debug-variables

You can use either the click classes or click ID if these are available and unique, or for an element like click URL we may select condition like contains “respond”.

That’s all! A quick way to properly set your click tracking.

P.S. Got a Google Analytics question? Send it to me and I’ll try to answer it on the blog.

I can’t accept Google Analytics Terms

You’re creating your new Google Analytics account. You fill in the fields, click on Get Tracking ID, the Terms of Service Agreement appears and… it seems there’s no way to accept it! Even if you scroll-down to the bottom, there’s no button.

No I Accept button

This has been a UI issue for quite some time now.

Here’s the quick work-around: if you’re using Chrome browser point your cursor to the header of the pop-up, it will change to show you can move it, just drag it up and you’ll see the two options: I Accept and I Do Not Accept.

Accept GA Terms

Now you can finish your sign up process and get your tracking code.

Happy Analytics!

P.S. Got a Google Analytics question? Send it to me and I’ll try to answer it on the blog.

Sampled data visualization changes in Google Analytics

Today I noticed a small change in the way Google Analytics show sampled data in one of my accounts. The well-known yellow rectangle is replaced by a permanent notice in the top right corner.

Update: This seems to be available only in accounts that contain a Google Analytics Premium property. Also, the issue is fixed an by default it takes 500,000 sessions.

Here’s how it looked before:

before1

And the slider to control the number of sessions and the precision:

before2

Now you can see for every report how many sessions were used for its calculation:

now1

And when you’re working with bigger websites and select a secondary dimension or apply an advanced segment, the data is sampled as usual:

now2

This new visualization could be easily missed without the alarming yellow color.

By default Google Analytics samples the data at 250,000 sessions, with a maximum of 500,000 sessions. I think this may be a temporary bug, but the drop-down that says Faster response, less precision and Slower response, greater precision is incorrectly marked as greater precision at 250,000 sessions:

now3

A work-around is to select Faster response, less precision first and then switch to Slower response, greater precision to get to the limit of 500,000 sessions:

now4
For specific ways to manage sampled data in Google Analytics you can check this great blog post at Luna Metrics.

Update: It seems GA fixed the issue and now you see the proper limitation for Slower response, greater precision option.

P.S. Got a Google Analytics question? Send it to me and I’ll try to answer it on the blog.