Posted by tian_wang
With today’s blog post I’m sharing everything one needs to know about an underappreciated tool: the Google Analytics add-on for Google Sheets. In this post I’ll be covering the following:
1. What is the Google Analytics add-on?
2. How to install and set up the Google Analytics add-on.
3. How to create a custom report with the Google Analytics add-on.
4. A step-by-step worked example of setting up an automated report.
5. Further considerations and pitfalls to avoid.
Thanks to Moz for having me, and for giving me the chance to write about this simple and powerful tool!
1. What is the Google Analytics add-on and why should I care?
I’m glad I asked. Simply put, the Google Analytics add-on is an extension for Google Sheets that allows you to create custom reports within Sheets. The add-on works by linking up to an existing Analytics account, using Google’s Analytics API and Regular Expressions to filter the data you want to pull, and finally gathering the data into an easy and intuitive format that’s ripe for reporting.
The Google Analytics add-on’s real value-add to a reporting workflow is that it’s extremely flexible, reliable, and a real time-saver. Your reporting will still be constrained by the limitations of Sheets itself (as compared to, say, Excel), but the Sheets framework has served almost every reporting need I’ve come across to date and the same will probably be true for most of you!
In a nutshell, the Add-On allows you to:
- Pull any data that you’d be able to access in the Analytics API (i.e analytics.google.com) directly into a spreadsheet
- Easily compare historical data across time periods
- Filter and segment your data
- Automate regular reporting
- Make tweaks to existing reports to get new data (no more re-inventing wheels!)
If this all sounds like you could use it, read on!
2. Getting started: How to install and set up the Google Analytics add-on
2A. Installing the Google Analytics add-on
- Go into Google Sheets.
- On the header bar, under your Workbook’s title, click add-on.
- This opens a drop-down menu — click “Get add-ons.”
- In the following window, type “Google Analytics" into the search bar on the top right and hit enter.
- The first result is the add-on we want, so go ahead and install it.
- Refresh your page and confirm the add-on is installed by clicking “Add-ons” again. You should see an option for “Google Analytics.”
That’s all there is to installation!
2B. Setting up the Google Analytics add-on
Now that we have the Google Analytics add-on installed, we need to set it up by linking it to an Analytics account before we can use it.
- Under the “Add-ons” tab in Sheets, hover “Google Analytics” to expose a side-bar as shown below.
- Click “Create New Report.” You’ll see a menu appear on the right side of your screen.
- In this menu, set the account information to the Analytics account you want to measure.
- Fill out the metrics and dimensions you want to analyze. You can further customize segmentation within the report itself later, so just choose a simple set for now.
- Click “Create Report.” The output will be a new sheet, with a report configuration that looks like this:
- Note: This is NOT your report. This is the setup configuration for you to let the add-on know exactly what information you’d like to see in the report.
Once you’ve arrived at this step, your set-up phase is done!
Next we’ll look at what these parameters mean, and how to customize them to tailor the data you receive.
3. Creating a custom report with the Google Analytics add-on
So now you have all these weird boxes and you’re probably wondering what you need to fill out and what you don’t.
Before we get into that, let’s take a look at what happens if you don’t fill out anything additional, and just run the report from here.
To run a configured report, click back into the “Add-Ons” menu and go to Google Analytics. From there, click “Run Reports.” Make sure you have your configuration sheet open when you do this!
You’ll get a notification that the report was either successfully created, or that something went wrong (this might require some troubleshooting).
Following the example above, your output will look something like this:
This is your actual report. Hooray! So what are we actually seeing? Let’s go back to the “Report Configuration” sheet to find out.
The report configuration:
Type and View ID are defaults that don’t need to be changed. Report Name is what you want your report to be called, and will be the name generated for the report sheet created when you run your reports.
So really, in the report configuration above, all the input we’re seeing is:
- Last N Days = 7
- Metrics = ga:users
In other words, this report shows the total number of sessions in the specified View ID over the last week. Interesting maybe, but not that helpful. Let’s see what happens if we make a few changes.
I’ve changed Last N Days from 7 to 30, and added Date as a Dimension. Running the report again yields the following output:
By increasing the range of data pulled from last 7 to 30 days, we get a data from a larger set of days. By adding date as a dimension, we can see how much traffic the site registered each day.
This is only scratching the surface of what the Google Analytics add-on can do. Here’s a breakdown of the parameters, and how to use them:
|
Parameter Name
|
Required?
|
Description & Notes
|
Example Value(s)
|
|
Report Name
|
No
|
The name of your report. This will be the name of the report sheet that's generated when you run reports. If you’re running multiple reports, and want to exclude one without deleting its configuration setup, delete the report name and the column will be ignored next time you run your reports.
|
“January Organic Traffic”
|
|
Type
|
No
|
Inputs are either “core” or “mcf,” representative of Google’s Core Reporting API and Multi-Channel Funnels API respectively. Core is the default and will serve most of your needs!
|
“core”
/
“mcf”
|
|
View (Profile) ID
|
Yes
|
The Analytics view that your report will pull data from. You can find your view ID in the Analytics interface, under the Admin tab.
|
ga:12345678
|
|
Start / End Date
|
No
|
Used alternatively with Last N Days (i.e you must use exactly one), allows you to specify a range of data to pull from.
|
2/1/2016 – 2/31/2016
|
|
Last N Days
|
No
|
Used alternatively with Start / End Date (i.e you must use exactly one), pulls data from the last N days from the current date. Counts backwards from the current date.
|
Any integer
|
|
Metrics
|
Yes
|
Metrics you want to pull. You can include multiple metrics per report. Documentation on Metrics and dimensions can be found in Google’s Metrics & Dimensions Explorer
|
“ga:sessions”
|
|
Dimensions
|
No
|
Dimensions you want your metrics to be segmented by. You can include multiple dimensions per report. Documentation on metrics and dimensions can be found here.
|
“ga:date”
|
|
Sort
|
No
|
Specifies an order to return your data by, can be used to organize data before generating a report. Note: you can only sort by metrics/dimensions that are included in your report.
|
“sort=ga:browser,
ga:country”
|
|
Filters
|
No
|
Filter the data included in your report based on any dimension (not just those included in the report).
|
“ga:country==Japan;
ga:sessions>5”
|
|
Segment
|
No
|
Use segments from the main reporting interface.
|
“users::condition::
ga:browser==Chrome”
|
|
Sampling Level
|
No
|
Directs the level of sampling for the data you’re pulling. Analytics samples data by default, but the add-on can increase the precision of sampling usage.
|
“HIGHER_PRECISION”
|
|
Start Index
|
No
|
Shows results starting from the current index (default = 1, not 0). For use with Max Results, when you want to retrieve paginated data (e.g if you’re pulling 2,000 results, and want to get results 1,001 – 2,000).
|
Integer
|
|
Max Results
|
No
|
Default is 1,000, can be raised to 10,000.
|
Integer up to 10,000
|
|
Spreadsheet URL
|
No
|
Sends your data to another spreadsheet.
|
URL for sheet where you want data to be sent
|
By using these parameters in concert, you can arrive at a customized report detailing exactly what you want. The best part is, once you’ve set up a report in your configuration sheet and confirmed the output is what you want, all you have to do to run it again is run your reports in the add-on! This makes regular reporting a breeze, while still bringing all the benefits of Sheets to bear.
Some important things to note and consider, when you’re setting up your configuration sheet:
- You can include multiple report configurations in the the sheet (see below):
In the image above, running the report configuration will produce four separate reports. You should NOT have one configuration sheet per report.
- Although you can have your reports generated in the same workbook as your configuration sheet, I recommend copying the data into another workbook or using the Spreadsheet URL parameter to do the same thing. Loading multiple reports in one workbook can create performance problems.
- You can schedule your reporting to run automatically by enabling scheduled reporting within the Google Analytics add-on. Note: this is only helpful if you are using “Last N Days” for your time parameter. If you’re using a date range, your report will just give you the same data for that range every month.
The regularity options are hourly, daily, weekly, and monthly.
4. Creating an automated report: A worked example
So now that we’ve installed, set up, and configured a report, next up is the big fish, the dream of anyone who’s had to do regular reporting: automation.
As an SEO, I use the Google Analytics add-on for this exact purpose for many of my clients. I’ll start by assuming you’ve installed and set up the add-on, and are ready to create a custom report configuration.
Step one: Outline a framework
Before we begin creating our report, it’s important we understand what we want to measure and how we want to measure it. For this example, let’s say we want to view organic traffic to a specific set of pages on our site from Chrome browsers and that we want to analyze the traffic month-over-month and year-over-year.
Step two: Understand your framework within the add-on
To get everything we want, we’ll use three separate reports: organic traffic in the past month (January 2016), organic traffic in the month before that (December 2015), and organic traffic in the past month, last year (January 2015). It’s possible to include this all in one report, but I recommend creating one report per date period, as it makes organizing your data and troubleshooting your configuration significantly easier.
Step three: Map your key elements to add-on parameters
Report One parameter breakdown:
Report Name – 1/1/2016
- Make it easily distinguishable from the other reports we’ll be running
Type – core
View (Profile) ID
- The account we want to pull data from
Start Date – 1/1/2016
- The beginning date we want to pull data from
End Date – 1/31/2016
- The cutoff date for the data we want to pull
Metrics – ga:sessions
- We want to analyze sessions for this report
Dimensions – ga:date
- Allows us to see traffic the site received each day in the specified range
Filters – ga:medium==organic;ga:landingpagepath=@resources
- We’ve included two filters, one that specifies only organic traffic and another that specifies sessions that had a landing page with “resources” in the URL (resources is the subdirectory on Distilled’s website that houses our editorial content)
- Properly filling out filters and segments requires specific syntax, which you can find on Google’s Core Reporting API resources.
Segments – sessions::condition::ga:browser==Chrome
- Specifies that we only want session data from Chrome browsers
Sampling Level – HIGHER_PRECISION
- Specifies that we want to minimize sampling for this data set
Report One output: Past month’s sessions
Now that we’ve set up our report, it’s time to run it and check the results.
So, in the month of January 2016, the resources section on Distilled’s website saw 10,365 sessions that satisfied the following conditions:
- organic source/medium
- landing page containing “resources”
- Chrome browser
But how do we know this is accurate? It’s impossible to tell at face value, but you can reliably check accuracy of a report by looking at the analogous view in Google Analytics itself.
Confirming Report One data
Since the Google Analytics add-on is an analogue to what you find on analytics.google.com, in your account, we can combine separate pieces in GA to achieve the same effect as our report:
Date Range
Organic Source/Medium
Landing Page Path & Browser
The result
Hooray!
Now that we’ve confirmed our framework works, and is showing us what we want, creating our other two reports can be done by simply copying the configuration and making minor adjustments to the parameters.
Since we want a month-over-month comparison and a year-over-year comparison for the exact same data, all we have to do is change the date range for the two reports.
One should detail the month before (December 2015) and the other should detail the same month in the previous year (January 2015). We can run these reports immediately.
The results?
Total Sessions In January 2015 (Reporting Month, Previous Year: 2,608
Total Sessions In December 2015 (Previous Month): 7,765
Total Sessions In January 2016 (Reporting Month): 10,365
We’re up 33% month-over-month and 297% year-over-year. Not bad!
Every month, we can update the dates in the configuration. For example, next month we’ll be examining February 2016, compared to January 2016 and February 2015. Constructing a dashboard can be done in Sheets, as well, by creating an additional sheet that references the outputs from your reports!
5. Closing observations and pitfalls to avoid
The Google Analytics add-on probably isn’t the perfect reporting solution that all digital marketers yearn for. When I first discovered the Google Analytics add-on for Google Sheets, I was intimidated by its use of Regular Expressions and thought that you needed to be a syntax savant to make full use of the tool. Since then, I haven’t become any better at Regular Expressions, but I’ve come to realize that the Google Analytics add-on is versatile enough that it can add value to most reporting processes, without the need for deep technical fluency.
I was able to cobble together each of the reports I needed by testing, breaking, and researching different combinations of segments, filters, and frameworks and I encourage you to do the same! You’ll most likely be able to arrive at the exact report you need, given enough time and patience.
One last thing to note: the Google Analytics interface (i.e what you use when you access your analytics account online) has built-in safeguards to ensure that the data you see matches the reporting level you’ve chosen. For example, if I click into a session-level report (e.g landing pages), I’ll see mostly session-level metrics. Similarly, clicking into a page-level report will return page-level metrics. In the Google Analytics add-on, however, this safeguard doesn’t exist due to the add-on being designed for greater versatility. It’s therefore all the more important that you’re thorough in outlining, designing, and building your reporting framework within the add-on. After you’ve configured a custom report and successfully run it, be sure to check your results against the Google Analytics interface!
Abraham Lincoln famously said, “Give me six hours to chop down a tree and I will spend the first four sharpening the axe.” Good advice in general that also holds true for using the Google Analytics add-on for Google Sheets.
Supplementary resource appendix:
- RegExr – General Regular Expressions resource.
- Debuggex – Visual Regular Expressions debugging tool.
Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!
from The Moz Blog http://ift.tt/1XUxxbz
via
IFTTT
Normally SEOs don’t set out to trigger a Google manual action unless they are specifically testing something. But Gary Illyes was asked an interesting question on Twitter. If a site had some kind of algorithmic penalty, would causing a manual action, then fixing it, cause the algorithmic issue to be lifted?
Last week we saw a sharp decline in the number of search results displaying the review stars rich snippets. Mozcast reported stars dropped from about 30% of search results down to 20%. However, a couple of days ago we saw those numbers start to jump up again. And now John Mueller has confirmed it was actually a bug.
While Joomla was rumored to be one of the CMS platforms that would be AMP ready, it seems that this still has not happoened. On the positive side, there now is a plugin option for sites that wish to implement AMP on Joomla without having to code it personally.
If you have implemented the official AMP plugin for WordPress, you have probably discovered that the plugin strips out the Google Analytics code from all AMP pages. But even though the plugin strips it out, AMP pages will still validate with Google Analytics code, but you need to add it manually.
