How to analyze data and create reports in Google Data Studio

Anna Shutko:

Hello, Michele, and welcome to the show.

Michele Kiss:

Hi Anna. Thanks for having me.

Anna Shutko:

Awesome. I’m really excited about today’s episode and I’m sure our listeners would love to learn more about Google Data Studio. So let’s begin. What is Google Data Studio? And maybe you could please tell us a bit more about what it is best suited for in case somebody doesn’t know.

Michele Kiss:

Sure. Yeah. I mean, I use it for a variety of different uses. I mean, essentially at its core, it’s a data visualization and a reporting tool where you can create fairly free reports so that you have a lot of say in how things are laid out and what kinds of visualizations that you include. So it can be used to build standard reporting that your stakeholders are going to look at every day, every week, every month.

But I also use it a lot for analysis because the ways that I can visualize the data are often a little bit more free and exploratory than the way that I might be able to if I were looking at say Google Analytics data in Google Analytics. So I both use it for building reports that are… Basically building standard reports. Building reports that users can engage with a little bit with some guardrails. And then also, I use it myself to explore the data and to actually do some analysis.

Anna Shutko:

Yeah. I think that’s a very important distinction that you just mentioned. So building a report and then actually using Google Data Studio for analysis. So if we talk about that a little bit more, could you please tell us what is the difference between marketing reporting and marketing analysis? Because like I mentioned, I think it’s very important for marketers to understand this distinction.

Michele Kiss:

Yeah. So the way I think of reporting versus analysis is that reporting is going to be or should, first of all, be derived from what your business requirements are, what things are important to your business. What do you need to be measuring to understand if you’re successful? And in addition, what data is going to support your understanding of that? So it may not be one of your key metrics, but it is something that ultimately factors into your understanding of your key metrics.

And so reporting is standard. It comes from requirements. It is a standardized format. You look at the same things every day, every week, every month, whatever that may be. Whereas analysis is going to be something that is much more free form. You can think of analysis as one little step past reporting.
So for example, if you’re giving a stakeholder the ability to do a little bit of analysis within your reports, they can pick some different filters and they can change some settings. And that allows them to explore the data a little bit. That’s kind of like this intermediate step. And then on the full-on analysis side, it’s just like you’re exploring some hypothesis that you have of the data and you’re kind of taking it down whatever route makes sense based on what you find. So it’s very much just an exploration of what the data can help you to answer as far as your business questions or your hypothesis about why something’s happening.

Anna Shutko:

Right. I think you made a really good point about the understanding of your data and analysis being more freeform. Yeah, I definitely do agree with this. And I also think that you really should try to create many comprehensive reports to try to fully analyze what’s happening with your key metrics. But if we talk more about the reporting part, can you please share what kinds of reports can marketers create with Data Studio more specifically? And in addition to this, what should you pay attention to when you’re creating different kinds of reports for different audiences? So the analysis part could be maybe tied more to your own analysis, but what if you’re creating a report for other people to analyze?

Michele Kiss:

Yeah. There’s definitely not a limited number of reports. You can have something that is a very high level that spans across say all of your business metrics. Or if you’re doing site analytics or app analytics, something that spans across the entire website, the entire app that pulls data together from, what traffic are you getting to what people actually do, what features they engage with, how they convert, all of that. You also though can have very specific reporting where all you are looking at is one marketing campaign, for example. And the entire report is just based on that one campaign and the things that are important for that one campaign. So for example, if the campaign has a specific goal that it wants to say drive users to a form and get them to submit an email address and get some gated content, then you could be looking at how well that campaign is doing at driving that. And the entire report could just be really specific to that one thing.

So I think that there’s a pretty wide variety. You can also, and it’s one of the things that I find really helpful, draw together data from a lot of different sources. So for example, I work with one client who has several apps. They have an Android app. They have an iOS app and they actually have a web app. So you can kind of engage with their product through a browser as well.

And so using Data Studio, we can pull together data from all of these different datasets and bring it together. The end-users ultimately don’t really know that they’re coming from different data sets. And we can look at these three different experiences and we can say, well, what features for example are important in each of them and how may they differ so that people could compare and contrast?

It’s so flexible in terms of what you can do, but it can be starting from just your key performance indicators and those critical metrics through to really specific about one experience.

Anna Shutko:

Right. Actually, I really, really liked your point about having a specific report for a specific campaign. And at the same time, you can have a report that combines data from multiple different sources. But if we go back to the process of report building and maybe analyzing data, can you give us a couple of tips? What should marketers do to create a report that helps you analyze data easily? Or maybe any kind of tips or what kind of visualizations you could use?

Michele Kiss:

Yeah. I mean, I have some rules that I follow as far as visualizations and they’re not terribly unique. Anything you do in Data Studio is going to follow the general rules of data visualization, which it should. So things like if what you’re trying to show users is a key performance indicator, you may use a scorecard metric to show your one big number. If you want to show them the trend of that, you’re going to use lines. If you want to show a comparison between values, you’re going to use a bar or a column chart. So it depends on each data point that you’re trying to include, I like to take a step back and say, “What do I want people to take away from this as I put them together?”

Michele Kiss:

I think also, it’s incredibly important to not overcrowd things. So I have things that I very rarely do. Typically, I won’t put unrelated data sets or unrelated data points together in a visualization unless there’s some reason for it. So a good example of that is if you’re looking at something like traffic and conversion right there, often, they don’t trend in the same way… They’re going to trend opposite. So when one goes up, one typically goes down. And so people get kind of confused if you shove them together on one chart.

So I tend to have sort of some ground rules about the ways that I will put things together. What I try to do, or what I think are really important roles of an analyst is with Data Studio is to set up the right guardrails. So an analyst understands a lot about the data. And if we take something like Google Analytics as an example, they’ll understand how the data is being captured, what dimensions and metrics you can put together, which ones you cannot put together. And just what makes logical sense in terms of how you can break data down?

And so that’s really the value that an analyst can add, is by creating visualizations with the right dimensions and metrics that make sense and filters that are appropriate for that. So you can create this little sub-world in your report that has users can filter by date, or they can filter by different devices, or different countries, or the channel that the user came in when they landed on your site. And you can create this safe space where they are able to engage with the data, but they’re not able to do something silly like pulling together things that make absolutely no sense. And I do think that that’s like a really important role, is to build things that are safely interactive.

Anna Shutko:

Yeah. I really like how you describe Data Studio as this kind of a playground where you can create a certain set of rules, but also at the same time, you can make it so that nothing spins out of control and people can actually use all these rules to their advantage. And also, another thing I was really curious about. So Data Studio added custom visualizations a couple of months back. Now, we talked about visualization, so would be great to learn more about what’d you think about custom visualizations? Are there any specific interesting things that the audience can learn about them? Are there any interesting ways in which you can use these custom visualizations in these reports?

Michele Kiss:

Yeah, I’ll confess. I don’t use a ton of customer community visualizations. I find that I tend to push the limits of the standard ones as much as possible for two reasons. One is perhaps a little bit historical. So because I was always working within these realms of needing these visualizations and making what was available in the standard ones work. So I feel like I found some really good ways to do that that are very effective and very simple rather than kind of adding this extra layer in. But the other thing is that as much as possible if I can convey data using a more standard visualization, I am going to choose to do that rather than something custom because it’s typically less mental overhead for the end-user. So if I’m using a standard visualization that they’re used to seeing, that’s completely normal. All they have to think about is how to interpret the data itself. They don’t have to think about how to even read the chart.

Now, if there are situations where sometimes the standard visualization is just not as good, it’s not as good a way to visualize it, then that’s going to be a different story. But I will actually confess that I tend to defer to the standards as much as I can where that makes the most sense. And there are two things I would say. So places that I use the standard ones in maybe not the typical ways or like using bar charts, for example, to create funnel visualizations. Because practically, every website needs some kind of reporting on that. Or even just things like how you use charts together. So using the color from one effectively as the legend from another and putting those together so that you don’t have to waste space on having a legend when you can use one chart to inform the other and also just have less clutter on your page. I think sometimes it’s those combinations of how you put things together that’s almost more valuable than the individual.

Anna Shutko:

Right. Yeah. It does sound very interesting. And I really like how you mentioned the use of color because I think it’s a very, very important part of any visualization. And I’ve also noticed some marketers do forget about this or they tend to misuse the color and actually apart from that. So you’ve mentioned a couple of tips on how you can use color visualizations on different other elements in Data Studio, but what are the typical mistakes marketers make when they create a Data Studio report?

Michele Kiss:

Well, if I’m going to be a flippant analytics person talking about anything reporting or data visualization, I would of course have to start with using pie charts. I feel like it’s not a conversation about anything related to data visualization when I say pie charts. I think the biggest mistake that I see is cramming too much in and not giving enough space for information to be digested. And it’s that idea of like, “Well, the report has to be one page. So by golly, I am going to shove everything onto that page.” And it makes it very difficult for a user to follow. It’s like basically anything that you do that would make a designer start sobbing about the lack of white space is probably not going to be a great choice. And so one is just trying to put too much in too smaller space in a way that doesn’t say, “Hey, I need this person reading it to be able to understand it. And so if I give them a little more breathing room to do that, I’m going to make it an easier user experience for them.”

Michele Kiss:

I also think that… So there’s kind of this philosophy of the data pixel ratio that basically says… in data visualization generally that says that the pixels that you’re using should be devoted to displaying data and not to extra stuff. And so I often see things like people putting really heavy borders and gridlines and legends where they’re not necessary. And pictures where they’re not necessary or not helpful, or too many giant logos where they’re not helping you to understand the data. And basically, the focus is not as much on the data itself. And so I think the most successful you can be, and the fewer mistakes you will make are when you just follow the general principles of data visualization. If somebody were going to say to me, “Should I read a book on Data Studio? Or should I read a book on data visualization?” I’d say data visualization. You could figure out what buttons to price in Data Studio pretty easily. But I think the biggest mistakes that I actually see are related to just not quite understanding the concepts of how to visualize data in an effective way.

Anna Shutko:

Right. And I really love how you mentioned the white space and I definitely think you should make designers talking about happening. So I really also recommend learning how to use white space. And now let’s talk a bit more about your use of Data Studio. So are there any workloads you could share? So how do you use Data Studio as a part of your day-to-day work, day-to-day routine?

Michele Kiss:

Yeah. So I use Data Studio a lot to the extent I am in it at least everyday type of thing. My ideal workflow, sometimes this doesn’t always happen because life happens and rush things happen. However, my preferred method is to kind of start from some kind of list of what a report or interactive report needs to have in it. And so I’ll often start in a spreadsheet or a Google Doc or something like that. I don’t write with those pen things, but otherwise. Basically putting notes down on paper of like, “I need to include these things. These are the most important things. These are kind of the secondary things and I need users to be able to engage with it in this way.” So that might be like, “These are the key metrics that I care about. These are supporting metrics. And these are the ways I want users to be able to filter the data.”

So I will try to start with that. That’s really based on the business requirements. The business requirement comes from the stakeholder like, “What do I want this reporting to address? What information do I want it to give me?” And so then I’m able to kind of translate that and say like, “This is the foundation of what the report should have in it.” I very commonly will actually whiteboard what a report should look like. So I have an eight-foot whiteboard in my home office and I would literally stand at my whiteboard and I will draw it out before I build anything. And then when I kind of go to start building it, I typically will also start from a template.

So most of the time for each of my clients, or the sites, or maybe I’ll customize one if it’s something specific. I will build out the template of what the reporting should look like. Things like: what are the colors that I’m going to be using? What are the fonts I’m using? How are the navigation and the layout going to work? Should there be borders? Should there be grid lines? All of that kind of information and having things like common logos and common headers. So basically, if you open up a report I’ve created, you’re like, “Oh, Michele created this.” It’s really clear because they all follow a very standard templated format. And that’s pretty intentional because I find it’s easier for people to always be looking at a similar format.

Once I sort of have that template, then that’s when I actually start building something. It’s definitely a little different if I am building it for analysis because typically, those are for me. And those are more like I just start throwing stuff down. And you know that feeling you have when somebody looks at your notes for something and you’re like, “Oh God.” It’s like you’re peering into my brain. It’s disorganized chaos. If it’s an analysis and I’m the only person looking at it, then yeah, I definitely jump straight to it. But if it’s something that I’m sharing with others, I do kind of like to go through this process of what the requirements are and then what I need to have in the report and then how it should look and then building the template and finally building the report itself.

Anna Shutko:

Wow. I think this whiteboard tip definitely sounds amazing. I have never heard of that. I’m definitely noting down that one. Michele, thank you so much for today’s conversation. I loved it. Where can the audience learn more about you?

Michele Kiss:

Yeah. All of the internet. I’m on LinkedIn, I’m on Twitter. I am also in a really amazing group. If people haven’t heard about it, it’s called Measure Chat. So it’s, oh God, at least 10,000 people in the community. It’s in a Slack instance and you can join it by going to join.measure.chat. But basically, it has a number of different channels. And one of them is a dedicated Data Studio channel. And there are some incredibly experienced Data Studio people in there. So you have the craziest question and people are like, “Oh, yeah, you need to do this and this and this.” So I’m pretty active within Measure Chat.

And then I do have a CXL course about Data Studio. So if you enjoy listening to me talk for four hours, apologies in advance, then that is available. And I’m actually also going to be talking about Data Studio at the upcoming Immerse Google Analytics Summit. It’s kind of the Dutch Google Analytics user conference. And I think all those videos are going live on March 4th. So I also talk a little bit about that and there are some visuals of some of the stuff that we talked about today actually.

Anna Shutko:

Michele, thank you so much for joining me today.

Michele Kiss:

Thank you so much, Anna.