Digital Analytics Scholarship @ CXL Institute — 6th Week Review

Google Data Studio Has been my tool of choice since the time I’ve known it. But it is a hard tool to master and no matter how much you know about it, there are still some new tips and tricks that you need to learn. This is what I think I’m about to learn in this course. Let’s see what is out there to learn about GDS.

6- Google Data Studio by Michele Kiss

• This course is focused on my beloved Google Data Studio and I am gonna like it.
• The tutor is named Michele and she has been learning GDS by trial and error. So she is the best tutor to teach such a great tool.
• One of the first things that Michele talked about was the differences between dimensions and metrics. The simple explanation is that you can do the math on the metrics. But dimensions are only text fields and can’t be measured. Dimensions are shown as green boxes and metrics are blue. These are the simple things you need to know about dimensions and metrics.
• And next, we have data sources. These are the data that you put in and use in your reports. They can be things like google analytics data or data from other Google services. It is actually great that you can input any kind of data into your google data studio reports and try to visualize data in a much better way.
• And you know what? You can make data sources report level to access them anytime from any chart.
• But first, you need to know that you have to have a mental picture of the things that you want to achieve and want them to be present in your visualization. If you use a pen and paper to draw the first draft of your mental picture, it’d be great and gives you a lot of insights about the final report.
• And now we are talking about all of the different charts that are available in the Google data studio. Charts are kinda the soul and heart of GDS and anybody who wants to be a master in it, needs to learn how to use them.
• We have a limited number of chart types in data studio and once you learn about them, all of the other ones suddenly become understandable to you. The first one of those is a line series. It shows the changes in a metric over time and you can track its trends in different time periods.
• After that, we have an area chart. It is solid and filled in the chart that I don’t exactly know how to use, but it is there for you to discover!
• Can we talk about charts and forget about bar/column charts? They are basically the same and can be used interchangeably. One is vertical and the other one is horizontal.
• We have another chart type which is very much like the line chart we had before. You can subtract time and axes from the time series chart and the result will be the sparkline chart. The best combination is to use it alongside scorecards, and the result will be very impressive.
• And yeah, we can use bar charts inside the tables. Instead of putting only numbers, you can actually show them using bars and heatmaps which is great.
• The most important thing is to use space real estate in the best way. Michele says that you have to pay attention to the data pixel ratio. The viewer should only see the most important data and nothing but it.
• Oh, and we have something called an optional metric in google data studio. It basically allows the viewer to add another metric to a table which is disabled by default and can be toggled on.
• I had this experience of using scorecards (single numbers) in rows and columns to replicate the feeling of a table. It is a bit time consuming, but the result is usually satisfying.
• Pie chart or donut charts? Which one do you like the most? Michelle actually doesn’t like them at all and thinks they almost every time confuse the viewers and divert their attention. But she says if you want to use pie/donut charts, try to limit the dimension items to something between 3–5.
• And we can utilize geographical data visualization in the google data studio. It gives us a big picture of the users we have and their actions based on their physical location.
• We are still talking about charts and this time, I want to talk about pivot tables. You might remember them from google sheets or Microsoft excel. But they are available in GDS as well. I have just understood the difference between pivot tables and standard tables. Standard tables consist of metrics in the columns and dimensions in the rows. But in pivot tables, we have dimensions both in rows and columns. The data in between is filled by any metric we prefer.
• But using only one chart to transfer meaningful data is a waste of time. If you want to be like a pro and make a powerful dashboard, you need to combine all of those different charts skillfully.
• You might want to rename metrics and dimensions to have a more meaningful presentation. But remember not to rename them in the data source itself. Instead, try to rename them at the chart level so that you can find the original source in the future.
• Michelle Kiss apparently doesn’t like to use legends at all or at least, want to use them as low as possible. So she tries to use shared legends, makes charts as legends, using scorecards as legends, and things like that. She is indeed a genius in working with GDS tools.
• And I learned about drilling down metrics. They will let you make something like google analytics reports in which the user can click on specific dimensions and see more granular data in the next step. It is a bit counter-intuitive, but very practical once you learn how to use them.
• Guess what is the topic of the next lesson? Colors! Michele says that almost 10 percent of all people are color-blind and you need to take this into consideration. If you overuse colors like green and red in your charts, you might be ignoring this part of society.
• Colors are very helpful in conveying different meanings and setting proper contexts. But you have to use them consistently. For example, if you used color for a specific metric, you shouldn’t be using another color for the same metric in another chart.
• Google data studio has a good option to color by dimension value. You once set the color for that metric or dimension and afterward each time that particular metric or dimension is used another place will be colored to that same color.
• We recently have got the chance to use conditional formatting in google data studio as well. I like the way this feature works in google sheets and being able to use it here the same way is really a gift.
• Data Studio and time are best friends. They need each other to show the best visualizations to the users. But the date has its own set of rules and limitations in data studio as well.
• I have the experience of placing some date controls in my data studio dashboards, but they don’t seem to be enough. The option of letting people compare different date ranges is something that will give your users more context in order to better understand whatever you are trying to show them.
• If you want to use date in your hand-built data sources, follow the lead of the data studio and use the format YYYYMMDD in those data sources.
• The next topic was the filters. They can be applied at the report level, page level, chart level, or group of charts level. Using filters will help to limit the source data and use it only for the purpose that you want.
• One other type of filtering data is using google analytics segments. Both system segments and custom segments can be used in the Google data studio to filter the data that google analytics has provided. It is great to use segments in GDS.
• But these two methods are not the only methods that you can use for filtering data. You have this option to give the end-user this permission to control the data herself. You just need to place a data controller in your reports and specify the metric and dimension that users can interact with. The rest is quite simple.
• You can also give the user the option to decide on the source of the data. For example, you can give him the chance to see data from google analytics or google search console data!
• Time is also a filter that can be leveraged in your reports. We have already talked about it in previous sections and it is there to be used in your reports.
• The next thing that you need to know about is calculated fields. They are somehow the same as custom metrics or custom dimensions. But in google data studio they are named this way.
• Using calculated fields, we can measure and show things that are not available in our data source, but we need them anyway. So we create them using calculated fields.
• In calculated fields, we can do math operations on metrics and text operations on dimensions. We can also use other metrics and dimensions to make new ones. But have to consider this: metrics and dimensions can’t and shouldn’t be mixed in a calculated field.it is the rule!
• In using calculated fields, we can leverage the power of functions like CASE which is a conditional statement for fields. We can also use regex-related functions like regexp_match or regexp_extract for creating new calculated fields. This is the power of Google data studio.
• Please notice that calculated fields can be used in report level and chart level. If you want the new field to be available everywhere, you have to create them at the report level. Otherwise, use the chart level method.
• The next topic is about data blending which is a huge one, considering its wide application in reporting. Google Data Studio uses the left join key method for blending data. It means that the data source in the far left is the basis of blending. If you want to be inclusive in your blended data, you have to place it in the far left position.
• Up to five data sources can be used in the data blending and a joining key should be set for blending. The joining key is something that exists in both data sources and connects those data sources together.
• There is a method called self to blend in which one data source gets blended with itself to help you make better visualizations. It seems a little odd in the beginning, but it is actually helpful once you understand how they really work.
• You can blend data by selecting two or more charts and right-click to find the blend data option. You can also first blend data in the sources section and then use them in new reports.
• When you want to share reports, you have to know that there are two types of sharing. One is the viewer’s credential and the other is the owner’s credential. If you want the viewer to view the report using the data source that you as the owner have access to, use the owner’s credential. But if you want the report to work based on the data that the viewer herself has access to, switch to the viewer’s credential.
• In Data Studio, we have this feature to automate reporting on timed periods. You can specify predefined periods at which specific reports will be sent to specific people.
• You can also export (download) reports to hand them to your boss or a colleague of yours.
• If you have the experience of working with google analytics, you should be familiar with sampling. Google data studio might sample your data in case of querying things. For example, when you apply a filter or set breakdown dimensions, your reports will be sampled.
• It is a good thing that the data studio will tell you if the data has been sampled. But if the data source is a blended one, you will not know if the data has been sampled.
• And the final things that Michelle talked about were a bunch of useful features. For example, if you need to add google analytics annotations to your GDS reports, she suggested that you copy and paste all of the annotations to a new google sheets file to be able to add it as a new data source and show them alongside the google analytics data. I think this was a cool thing.
• The other thing that was mentioned was about showcasing funnel steps. To be honest, I didn’t truly understand what she did there and don’t know what she was really after. But there were a bunch of calculated fields and placing them in the side by side reports.
• And finally, Michelle mentioned a Slack Group that all google products related topics are being discussed and almost everybody in digital analytics needs to join.