Start learning with our library of video tutorials taught by experts. Get started
Viewers: in countries Watching now:
In Google Analytics Essential Training, Corey Koberg shows how to use the Google web analytics platform to generate and evaluate information about the visitors to a web site, including data on site traffic, user behavior, and marketing effectiveness. This course covers the out-of-the-box functionality, from account creation to reporting fundamentals, and explains how to glean insights from the vast array of data available.
Inline filters are a simple but powerful tool to allow us to quickly control and consolidate the data that we're analyzing in the data table and its graphs. Before we begin, let's talk about terminology. There are three primary types of filters; profile filters, inline filters, and advanced filters. If we take a look at the profile settings in the Filters tab, here we'll see some advanced profile filters that will restrict some of the data that we can get into the profiles. When we talk about inline or advanced filters, we're not talking about profile filters at all. Rather, we're talking about the filters at the top of the data table and the reports.
Let's take a look. As we move here to the Languages report, in the Visitors section, under Demographics, we can see all the different languages of people who are visiting our site. If I was interested in analyzing just the Spanish language visitors, I could click on es. The problem is that different browsers report Spanish in different ways, and I want to capture all the Spanish language users. So what I can do is come here to the filter box at the top of the table, and I can click on es, click the magnifying glass to run the filter, and this will capture all the different lines that contain es, whether it's just by itself, whether it's Spanish, whether it's going to be a Latin American, Caribbean, Mexican, etcetera.
From here, I'm able to evaluate them individually, and see how many visits came from each different version, but I can also see what the group represents as a whole. On the scorecard across the top, I can see the metrics for all of them summed together, and the data over time graph updates as well. So I can see here, there were 20,000 visits in total from Spanish users on my site, who stayed for an average of 2.9 pages per visit, 1 minute and 6 seconds, 92% of them were new visits, etcetera. There are tons of uses for this type of filtering inside these reports. Let's take a look at some more. Click on over to the All Traffic report inside the Traffic Sources tab.
Here, we see the different sources and mediums that bring traffic to the site. Maybe your question is, how many blogspot blogs are bringing traffic to your site? If we simply type in blogspot here in the filter box, hit Return, and what we'll see down here is a list of all the different blogspot blogs that are bringing traffic to my site, and the scorecard, and the data over time graph will update to reflect the overall aggregate numbers for those blogspots put together. If I am interested in individual blogspot blogs, I can look down here and see all the different metrics associated with each one of those blogs. Another possible example is, what if we were interested in the number of Apple devices sending traffic to our site? If I click on Visitors, and Mobile Devices, we can filter for Apple.
Type Apple into the search box, and what we're going to see is the data table down here is restricted to only things that contain Apple, and we can see individually which Apple products are bringing them, as well as the total amount that Apple is bring in to my site across the top. Let's head over to the organic search report to see what else we can do. This report under Traffic Sources > Sources > Search, and Organic is going to default to show the keywords that users search to find our site. When we talk about keywords that bring people to my site, we can break these down into two different kinds of keywords; keywords that indicate someone is already familiar with my site, my brand, or my company, and the more generic keywords, where someone was not necessarily familiar with my company at all, but was simply looking for some information, or a solution to their problem.
For example, if you typed in the word Google, or any derivative of that, you're probably already familiar with the company Google; you're not just looking for a generic search engine. Now, if you type in something like Google merchandise, you might not be looking for the Google Store specifically, but just a store that happens to sell Google stuff. In fact, you may not know such a thing as the Google Store even exists. When we're working on SEO for your site, most people tend to want to focus on these non-branded terms. These are the terms that are going to bring you net new visitors, and let's face it, if you're not ranking for your own name, then your SEO has other issues.
When we're ranking for some of these generic keywords, it can take a bit more effort, and we want to measure our progress using Google Analytics. To do this, we'll go over here next to the filter box, and we're going to click Advanced. This is going to drop down some more options to filter our data. For our purposes, we want to exclude keywords containing our brand name, Google, and click Apply. So we type in google, we're going to change Include to Exclude Keyword, and click Apply. Now, I've definitely got rid of my branded term of Google, but I've got a few more terms in this list that would qualify as branded terms, or people who are familiar with my brand.
For example, I see YouTube in here, in different spelling variations; I see Android. Those are things that have to do with my brand as well. So let's go back to the Advanced filter, and we have this ability to run multiple filters on here. We have an and operator. We can click this again; we can select a dimension of Keyword. I can choose to Exclude, and I can put in words containing things like -- now we have YouTube; we have some people who spelled YouTube with a space, and others, So I am just going to put tube in here in general, and that should cover most of those. Click Apply. Okay, now we look down through the list, and we still see some Android ones I need to take care of. We see Google misspelled. In fact, we even see Google up here in Cyrillic.
So one thing to not is that this is not going to do any type of translation, or cover languages. If you specifically want to include foreign language versions of this, then you're going to have to address those individually. I can come up here, and I can continue to add and ones down here. I can also change this to do something else; for those of you that are familiar with regular expressions, I can use the vertical pipe bar here to add additional words that I want to exclude. So in this case, I want tube, and gogle, and android, and we are going to apply that. I am going to increase the number of rows to 50, and we can look down through list, and look for a few other variations of ones with more Os, ones without the O; there's plenty of misspellings in here. We can look for Chrome, and other ones through here.
I can continue to add these through here, and build up my list of branded keywords that I want to exclude. Now, one important piece to note here for SEO purposes is this second result here: not provided. In 2011, Google decided that for any user who is logged into their Google account -- perhaps they logged in from Gmail, or a Google Plus account, and never logged out -- that they will automatically be redirected to a secure version of the Google search engine whenever they search. When this happens, the Google search engine will not pass individual search terms through to Web analyst tools like Google Analytics.
It's an unfortunate loss of data, but everyone faces the same challenge, as no tools can recover the data, even Google analytics. So it's not a branded term, but it may skew your data. So if we want to filter that one out, let's go back to edit our filter again. We're going to add a dimension, select Keyword, Exclude, Exactly matching, and type in not provided. Apply this, and we can see that our list can update to where the not provided keywords will not be included in our list.
Okay, so if we continue on in this fashion, we can weed out our branded terms, and continue to improve our list, but let's go ahead and take this list of non-branded search terms, and sort them by performance. Now, bounce rate is a good metric, so let's look for ones that have a low bounce rate. We can easily find this by sorting on the Bounce Rate column. We click on the Bounce Rate column; first it's going to show us the largest bounce rate. Click it again, and we'll see bounce rate sorted by the least. So this is what we want; terms that don't bounce at all.
Now, what we've got here is a list of non -branded terms that have a great bounce rate, and so we all know that they're very valuable; we should pour all of our marketing efforts into this list, right? Well, not quite. As you astute viewers have noticed, this is actually not valuable data at all. In fact, this is nearly junk data. Just because one person typed product.asp into here doesn't mean that this is the hot new keyword, and I should run out and tell my CPC folks to go nuts on that as a new keyword. When we talk about performance, we're really talking about performance in the context of a reasonable amount of visitors.
So let's use filters to turn that English, a reasonable amount of visitors, into data that we can put into Google Analytics. What I really want to see are just the ones that brought in, let's say, five or more visits; not these onesy, twosy ones that are simply anomalies. We are going to go back into our Advanced filter; we're going to add another row. We're going to select a metric, and we're going to select Visits, and we're going to say Greater than 4, so that will give us 5 or more visits. Click Apply, and now we can see this is going to update to show us ones that still have a good bounce rate, but have a reasonable amount of visits.
This list looks much better. You see we've now got some real contenders, and with a bit more massaging, we'll have a true list of our best keywords. This type of data filtering is essential when you're dealing with any kind of rate based metric, such as e-commerce conversion rate, goal conversion rate, bounce rate, etcetera. Another tip for this type of analysis is to expand your date range to get as much keyword data as possible. Advanced filters and inline filters are incredibly powerful tools you'll use over and over again in your analysis. You'll find advanced filters particularly powerful anytime you're sorting by a rate to strip out those cases with just a handful of visits.
Find answers to the most frequently asked questions about Google Analytics Essential Training (2010).
Here are the FAQs that matched your search "":
Sorry, there are no matches for your search ""—to search again, type in another word or phrase and click search.
Access exercise files from a button right under the course name.
Search within course videos and transcripts, and jump right to the results.
Remove icons showing you already watched videos if you want to start over.
Make the video wide, narrow, full-screen, or pop the player out of the page into its own window.
Click on text in the transcript to jump to that spot in the video. As the video plays, the relevant spot in the transcript will be highlighted.