In this article, Julien Coquet lists ways to extract Google Analytics data or exploit data from your favourite analytics platform in ways other than exporting plain PDFs.
So you are tracking your site or app in Google Analytics. Or maybe you are using Adobe or Webtrends or AT Internet or Webtrekk or Piwik and sundry.You are now enjoying reports in a user-friendly interface. Assuming you configured your tracking correctly, that is. Hurray for data!
But the reporting interface only goes so far. It will, in the end, generate frustration for you as a digital marketer. Because you need to use the raw data from the reports.
In this post, we’re going to list ways to extract and exploit data from your Google Analytics or your favourite analytics platform. In ways other than exporting plain PDFs. The various techniques become more sophisticated as we progress down this post.
Depending on your organisation’s analytics maturity level, you will be sharing insights with your team and others. So you need to provide insights in a format they will understand. Just sending out a PDF curves and graphs is not going to cut it. Page views and sessions are great vanity metrics but what kind of story do they tell? What does a geo/map report say about the state of conversion by segments?
That’s right, you’ll need to get that data out of the platform. You need to rework it somehow. So you can display it in ways that allow you to effectively work with it. And it should be in a format that your “higher ups” will understand.
Low-hanging fruit: using built-in Google Analytics data extract functions
There are some simple steps to extract Google Analytics Data. In the Google Analytics interface, just about any report supports an “Export” button. You can often find it above the graphs and data tables. From there you can select a handful of export formats.
- looks “pretty”;
- easy to view across platforms;
- somewhat searchable;
- easy to print at scale
- rendered format, meaning it cannot be modified;
- data can’t be extracted
- Actual numbers you can use and import into your favourite data crunching solutions, although many of them have now developed APIs that connect go Google Analytics’.
- Ugly as heck. Then again, they’re just numbers 🙂
- Used by entire industries to share data;
- data can be integrated into other sheets;
- formulae can help
- Excel is still very clunky, despite dozens of versions;
- Number formatting can be hell for international (non-US) users.
- Leverage the Google ecosystem;
- awesome sharing/collaborative capabilities;
- *almost* functionally comparable to Excel
- *almost* functionally comparable to Excel;
- hard to access in Google-unfriendly companies and environments.
Great, now you know that you can extract data from Google Analytics. And in enough formats that you can actually use them on a daily basis. But as you can imagine, the goal of posts such as this one is to broaden your horizons and show you alternatives.
Let’s move on to advanced methods!
Advanced extraction: using the Reporting API
(aka the golden apple)
An API (Application Programming Interface) is a way for developers you access your applications. You can then query its data in both ‘read’ and ‘write’ mode.
The Reporting API arrived a bit late in Google Analytics’ evolution. It still is catching up to the platform’s features. But I for one use it on a daily basis. I’m either using the code I wrote/borrowed. Or I’m using a visualisation platform such as Domo or Tableau Software or others.
The API allows you to pull just about any dimension/metric in Google Analytics. With somewhat limited sampling
R / scripts:
- Leverage the power of the API and use the data crunching/processing functions inherent to said scripts (Python most notably) and R of course.
- Requires coding as well as decent (read: advanced) knowledge of statistics
BigQuery (Premium/360 only):
BigQuery is a subcategory of its own, as part of the Google Cloud Platform. It is reserved for Google Analytics 360 (née Premium) customers. Please hold comments about how much it costs, we know…!
- API extraction on steroids;
- essentially SQL for the full Google Analytics data set, and then some; full granular data;
- no sampling
- Requires Premium/360
Data visualisation platforms:
(Domo, Tableau Software, Klipfolio, BIME et al.)
- They create pretty graphs and dashboards that can help you make sense of your data
- dashboard sharing capabilities aren’t always adapted to corporate policies
On the topic of data visualisation, give Google’s Data Studio a try. On the topic of data visualisation, give Google’s Data Studio a try. It will allow you to create customizable, shareable dashboards. Based on your Google Analytics data. This will be available outside of the US “soon”.
Once listed, what data export/extract methods do I choose?
Finding your favourite/most useful export method is not easy. It boils down to how much slicing and dicing you want to apply to the data. Both before or after it is extracted from the interface or the API.Long story short, you need to check for obvious parameters:
- Dimension/column formatting
- Segment management
- Export size (batch exports, paging and report offsets)
- Time management (time stamps, time zone offsets)
- Date range comparisons
As I mentioned before, your reporting sophistication will vary, based on the following criteria:
- How your KPIs are defined for your company
- Your mastery of segmentation techniques
- How the data you extract will tie with other data sources
- How to mature your organisation is to act on data
- How much time and resources you or your team are willing to spend out of your already busy work week to generate reporting
In closing: extract your data from Google Analytics
In this post, I listed ways for you to extract useful & insightful out of Google Analytics. The techniques I mention here can, for the most part, apply to other analytics solutions. Google Analytics is likely the solution with which you’ll have the most immediate or prolonged exposure.
Before you start trolling in the comments: YES. Using a premium solution such as Google Analytics 360 can provide more extraction. And processing options such as BigQuery or Data Studio 360 as well. But let’s assume that us mere mortals cannot all get our hands on such sophistication ;-).
But what about you? How do you use your analytics data? Do you work on it outside of the reporting interface to conduct analyses? Let us know in the comments!