If you are using Google Analytics, you have access to a wealth of information—over 100 reports with over 350 dimensions and metrics! Yet inevitably, you also have a ton of data that lives outside of GA. It may be in a CRM system, content management platform, inventory database, or in other tools and systems that you use to run your business.
Wouldn’t it be great if you could integrate data from all these offline sources with the online data collected by Google Analytics?
Look no further than GA’s new Data Import feature! Data Import lets you upload your own data directly into GA to create a more complete picture of your business and produce more accurate analysis and segmentation.
Let’s take a look at some examples of why you may want to use Data Import, the types of data that you can upload, and some important things to watch out for along the way.
To import or not to import?
Just because you have a mountain of data available, importing it all into Google Analytics doesn’t automatically make it useful. Before you decide whether or not to import data, it’s important to ask yourself the following questions:
- Define your objective: What would I like to further understand about my customers or my business? What questions am I trying to answer?
- Assess your inventory of data: What data do I have available outside of GA that would help answer these questions?
- Identify your import key: Does my data have a common ID value that I can use to join GA data with my offline data sets? e.g. User-ID, Page URL, Product SKU
If you have a CRM system and have classified your customers according to specific personas or audience segments, perhaps you would like to use these segments for analysis in GA. With Data Import, you can attach custom dimensions to each person based on their User-ID.
If you have a blog, perhaps you want to know which author’s posts generate the most traffic. In GA, you may find only the title and URL of each article. But your content management system likely contains the author, category, date, and other characteristics of each post. You can then join this article “metadata” with the existing article dimensions in GA, using the Page URL as the key.
If you have an online apparel store, you may want to know which colours and sizes of your products are the most popular among different segments of users. But what if your GA ecommerce data only contains the product SKU? In this case, you can upload additional dimensions, like colour, size, and product name, for every SKU.
Data Import was announced at the 2014 Google Analytics Summit and takes Dimension Widening to a whole new level. In other words, Dimension Widening is now even wider! In particular, Data Import adds the ability to widen campaign dimensions, as well as product dimensions and metrics.
What types of data can I import?
User Data: Enable enhanced segmentation by uploading additional data about your customers, e.g. customer lifetime value, subscription status, audience segment. User Data import requires either a User-ID or user-scope custom dimension as the key. Beware that you may not upload Personally Identifiable Information (PII).
Campaign Data: Simplify your campaign tagging by using a single campaign parameter (e.g. utm_id) and importing the campaign name, source, medium, and other dimensions. Also, if you are using campaign tagging for a non-GA analytics tool (e.g. SiteCatalyst), you can now leverage those same campaign codes for GA.
Content Data: Enable better grouping of content by importing information about your articles or blog posts, e.g. author, topic category, date published. You can use either the Page URL or another hit-scope custom dimension as the schema key.
Cost Data: Gain a more complete picture of your paid advertising by importing data from non-Google campaigns. Note that this replaces the previous API-based Cost Data Upload. If you are using the existing cost data upload, take a look at this cost data migration guide.
Product Data: Enable more detailed analysis by importing additional information for each product SKU, e.g. size, colour, style. Product Data import requires that you are using Enhanced Ecommerce (ec.js).
Refund Data: Reconcile your internal ecommerce reporting with GA by importing refund data. Beware that refund data cannot be deleted or modified once it has been uploaded to GA. Refund Data import requires that you are using Enhanced Ecommerce (ec.js).
Custom Data: If none of the above options suit your needs, you can create a completely custom data set using a variety of GA dimensions and metrics. Currently, you cannot use custom variables, time-based dimensions (hour, minute, etc.), or geo-dimensions (country, city, etc.). The dimensions and metrics available for upload will also be limited by the scope (user, session, or hit) of your selected key dimension/metric.
Can I use Data Import instead of sending data with hits?
You may find it advantageous to use Data Import instead of appending additional data to GA hits. Here are some reasons why:
- Adding additional data to hits requires a developer to write custom code. Data Import can be managed through the GA admin interface.
- The data may not be available when the hit is sent. Data Import lets you upload data in advance so that it is available when the associated hits occur. (Query Time mode will also allow imported data to be applied retroactively.)
- Appending a large amount of data to each hit could slow down your site. With Data Import, you can send only the minimum amount of data with the hit and upload everything else.
In many cases, it will make sense to use a hybrid solution of data collection and data import. In this way, you need only collect a “key” with each hit (e.g. Page URL, User-ID, Product SKU, or Campaign ID) and all other associated dimensions can be uploaded and managed via Data Import.
Does Data Import apply to historical data?
Not yet, but it’s coming soon!
At the moment, Data Import is available in “Processing Time” mode. This means that your imported dimensions and metrics are joined only when new data is processed by GA. So, your imported data only applies going forward.
However, Google has also announced that a Query Time mode will be supported. This will allow your imported data to be applied whenever you generate a report. Hence, it will join imported data with the historical data already in GA. For now, Query Time is only available to GA Premium customers.
What else do I need to know?
- Some features of the new Data Import have not yet been rolled out to all users. If you would like to join the beta for full-access, sign-up using this form.
- For more information on Data Import, check out this Analytics Help Centre article and this announcement on the GA Blog.
Have you used Data Import yet? Tell us about it!