Google Data Studio - A Review
As a data analyst, sharing the story with data is the most important part of the job. That is why when Google announced Google Data Studio, I was excited to see how Google’s data visualization product would compete against the likes of tools from Tableau, IBM, Microstrategy, Microsoft, and other companies in this space.
What is it?
Introduced in August 2016, Google Data Studio (GDS) is a lightweight web-based data visualization tool with core functionality in visual interpretation of web analytics such as Google Ads and Google Analytics programs. GDS is posed to be a cost-effective (i.e. free) way to solve 80% of the reporting and analytics needs for enterprises carried out in a day-to-day fashion. GDS is still in development mode with regular releases and feature addition supported with an active community.
Simple Setup
How does GDS work? Similar to other visualization tools, one connects to ETL databases and/or ad hoc datasets (via data sources) in GDS, performs any schema modification if needed (e.g. adjust field type, add calculated fields), and creates visually compelling reports to tell a story. From my experience, the data preparation steps are straightforward and the user interface is clear and user-friendly.
GDS supports basic chart types that appeal to a broad audience, including line chart, time series, bar, pie, map, bullet charts as well as specialized visualization like the pivot heatmap chart. The chart types are not far from what other BI (Business Intelligence) tools provide, but the flexibility and the level at which you can fine-tune with GDS charts is more limited.
Here is a table of all supported charts in GDS:
Many Data Connectors
Currently, there are 18 native data source connectors (as of this post is written), from Google Analytics and Google Ads to PostgreSQL, as well as partnered data connectors to bring in other pieces of data in Facebook, Amazon or elsewhere. Where GDS really stands out is as a natural integration with other G-suite tools - when you connect to Google Analytics for web analytics data, GDS automatically knows how to interpret the metrics correctly and tags will be product names rather than meaningless character strings. In other BI tools, this requires additional data preparation.
As an example, connecting to Google Analytics to retrieve data follows steps like these:
However, I encountered an issue when I added a new data source for a Google Ads account with campaign data. The data connection was established successfully but the report failed to load any data to the visualization report. The less satisfying part was that there wasn’t much error message or troubleshooting tips. It was a bit confusing and I suggest fixes should be prioritized at Google.
Easy Collaboration and Sharing
What about other important aspects of data visualization tools? Visualization tools are built for sharing and collaboration within or across the organization. GDS is admirable on shareability - view- and edit- permissions are assigned based on Google accounts and reports exist independent of the underlying data. In other words, viewers won’t be able to alter the original data without separate access granted. You can share the report by URL with the audience, schedule daily email delivery, download the report as a PDF file, and even embed the report as part of a webpage. This is the distinguishing feature for GDS that leaves other BI tools behind.
(The upper-right tab conveniently organizes all functions for collaboration in one place)
There is a huge advantage of using GDS, in that there is no overhead cost or a significant investment in time to enable group access and allowing the visualizations to be more widely used. In comparison, Tableau’s freemium version publishes all report to the open web without permission control.
In terms of computation performance and scalability, GDS is not optimized yet. It’s been reported that queries will run 30 secs before charts are fully loaded so you won’t want to feed in large datasets and hope to see updates in seconds. The performance also vary by data source. With some data sources (including Google Ads, Analytics, sheets, and Cloud SQL) GDS analyzes the dimensions, metrics and filters contained in the report, predicts possible queries, execute these queries in the background and stores the responses in the predictive cache, so the queries should return faster than other data connections. Predictive cache is a common practice in BI tools, like Tableau, and I’m glad to see GDS has an implementation as well.
Limited Customization
GDS has a few major drawbacks though. There is little to no ability to modify the underlying data, fewer than 50 built-in functions for calculation and a limited number of charting options than other similar tools. For example, the funnel visualization chart (or any flow charts) is a strongly sought-after feature for marketers, which is currently not available in GDS. Also, there is a lack of connectivity between GDS and popular data repositories, including Amazon Redshift, Cloudera Hadoop Hive, MS SQL Server, and Snowflake, just to name a few. The user interface design could be more flexible, for instance: if you have field names that are longer and more descriptive, they can be missed in the report. it’s not easily seen, as in the example below.
(In GDS, the editing panel has fixed width and long field names are left out (…))
(Below: Adjustable panel with to accommodate long field names in Tableau)
Better Tutorials
One of the things about GDS that received lots of complaints is how terrible the tutorials are. The tutorials created by Google were outdated (created in 2016) and
illustrated topics by letting learners click on the in-browser corresponding icon or answer in serial steps. Videos would be more intuitive for new users and they can come back and replay anytime they want. The tutorials only cover the basics and it will be useful to make additional tutorials available about connecting to individual Google products, for example, Google Analytics, Google Ads, and Google Sheet.
When to Use Data Studio
In my opinion, GDS is a pleasant addition to the visualization and reporting tooling space because I can see it shines in certain scenarios. Here are a few:
Scenario 1: Your organization uses Google suite of tools already.
If you’re working in marketing analytics and your organization already is a Google shop, i.e. Gmail, Google Ads, Google Analytics, and Google BigQuery are the ecosystem you’re working in, and your day-to-day tasks don’t involve complicated analytics (e.g. projections), GDS would be a strong contender for you that optimizes shareability, native integration with G-suite tools, and fast development for new dashboard or report, all with significant lower overhead costs compared to other BI tools.
Scenario 2: Your audience and user base is very broad.
If you have a wide target audience for your reporting tool and accessibility is a big factor to consider for your business, I would recommend trying out GDS since it’s 100% web-based, free, easy to use and with controlled authentication. While there are other free web-based analytical tools out there, none of them makes authentication and permissions as easy as GDS’s leveraging of pervasive Google accounts. The presumption is that you’re willing to work within some limitations of GDS.
Scenario 3: You want quick insights with low setup and ongoing costs.
If you’re a small online retail business owner and are already using website traffic data to bring business insights, I would recommend adding GDS into your toolset. Google Analytics is a powerful reporting tool. The power also happens to be a double-edged sword - there are too many pages and it usually takes a few clicks to go to your regularly visited pages. In GDS, you can create dashboards with the most important metrics for “you”, not for any regular user, combined with additional business data in sales or inventory to make business decisions. You can also create customized lists with goals like “Pages Need Affiliate Marketing” or “Pages with Best Affiliate Marketing” to remind yourself where to devote resources. Also, if you have a small budget for maintaining your visualization and reports, GDS is considered easier to make adjustments than some fine-tuned BI reports.
Overall
To sum up, GDS is a basic reporting tool that would serve a broad audience featuring user-friendly interface recognizable to anyone who has worked with Google tools and simple permission controls. The greatest strength of GDS comes when interpreting Google-sourced data from web analytics and databases to draw your own insights and visually represent important metrics to the audience. It’s free and has few competitors in the same space offering equivalent functionality and community of users. I’d say GDS is worth a try for everyone.
Try it here https://marketingplatform.google.com/about/data-studio/