Experiences With Tableau

[Zentangle]

Hey all, as we have passed Week 1, I am reaching out to share my experiences using Tableau for the first time, as well as asking other teams their experiences. To preface this, it is not a suggestion nor a advertisement, but a presentation of my experiences to help teams decide whether they would want to use it or not. All the following data was procured from my team, 4239, and our scouts during the 2020 Great Northern Regional. Data may not be 100% accurate. Let’s get right into it.

—PROS—

• Quick, easy data filtering and sorting
• Access across multiple different platforms and devices
• Live pulls from your database (we used google forms linked to a spreadsheet)
• New data comparisons and sheets can be made from any device (that is signed into an account with permissions to do so)
• Lots and lots of options to sort, as well as calculated fields for find average scores and such

—CONS—

• Limited by your data/WiFi connection quality (as @Anthony_Galea said, this applies only if you use the Live feature, otherwise you just need to import new data each time)
• Very slow on updating data (often 3-4 matches behind without manual updating)
• New sheets or data comparisons require a separate sign-in in your browser, outside of the app.
• To view, you must be signed in to everything it is going through (Google Drive, and Sheets, and Forms)
• Not very intuitive

------------THE DATA------------

To make this clearer, this is the average Power Cell points per match. This required us to scout the Low, Outer, and Inner port scores per robot, and then use a calculated field to multiply each value by its point value.

The teams that appear to be very inaccurate, yet have low seconds per ball are most likely low goal bots. This will be fixed for our next competition, omitting robots that didn’t score in the high goal at all.

This chart laid out the perceived effectiveness over the day of each team. However, this is objective data and next competition will have this chart using the total score per match instead of inherently variable objective data.

This chart is especially useful to determine the locations and timing of scored balls (In Auto, signified by A, and Tele-Op, signified by T) I found it useful to color-coordinate the scored balls in Red tints for Tele-op, and Blue tints for Auto, with lighter shades being the low goal.

Overall, it was a nice experience with a bit of learning curve. Admittedly, I didn’t allow near enough time to practice and get used to the program, yet over the next four weeks I should have plenty of time. I would like to hear from other teams that used Tableau and their experiences with the program.

Dylan Larson, 4239

[Anthony_Galea]

We ran Tableau completely offline at our event, it just required importing a new data source every time we wanted a data update. Worked really well even if we did not get to use our data in alliance selection.

[Bob_The_All_Mighty]

how does the quality of graphs n such compare to those google sheets can natively produce?
how do they compare in easy to use and create?
and do you think the step up in quality is worth it ?
im interested in trying out tableau for our scouting system for north star.

[Kaitlynmm569]

Im going to jump in here because my team used Tableau for the first time at Palmetto last weekend.

---

@Bob_The_All_Mighty

how do they compare in easy to use and create?

The issue I have with google sheets is that you need to know what data to put where. I found it very difficult in the past to create a chart exactly how I wanted. I was able to do it, but found it difficult. It’s not just as simple as choosing a parameter and putting it in a graph. In addition to this, filtering is not easy. If you want a graph with specific values (from a specific event, for example), you need to either add a filter to your existing master sheet, or create a new sheet with a =FILTER() function. With Tableau, this is not needed. In addition to this, you cannot easily change a filter parameter in GSheets if youre using a formula (or even if youre not) (I’ll elaborate below).

Don’t get me wrong, I struggled with Tableau too. Tableau has a bit of a learning curve in order to know how to effectively use it. I struggled with the difference between Measures and Dimensions and figuring out when to use them and where in order to get the type of the graph I wanted. The good thing about Tableau is that once you know how to use it, you know how to use it. Once I became familiar with all the buttons and functions, I was able to easily create graphs and charts looking at any data point I wanted, in a matter of minutes. Stories and Dashboards are also extremely helpful when wanting to view data quickly and easily, as well as sharing it with your team on a big screen (I love Stories for this). Going back to filtering, it is worlds easier with Tableau. While there is a need for separate sheets (you need sheets to build any graphs in Tableau), there is no need for separate formulas. To apply a filter, you simply need to drag that Dimension or Measure into the “Filters” box on the left. From there, you can add the filter to be visible on the panel on the right, and bam. Quick and easy way to change a filter without needing to mess with anything too drastic. Another bonus to filtering in Tableau, is that you can apply one sheet’s filter, to multiple sheets. This helps to minimize clutter on a dashboard or story.

@Bob_The_All_Mighty

how does the quality of graphs n such compare to those google sheets can natively produce?

Overall, the basic bar and line graphs are very similar. The bonus with Tableau is that you have the ability to have bigger more complicated graphs, such as maps, treemaps, heat maps, and plenty more. You have a lot more flexibility with Tableau graphs than you do GSheets. Another thing with Tableau that could be seen as either a pro or a con depending on how you look at it, is that you need certain types of data for certain graphs. For example, for anything that includes a line graph or an area graph, you need one Date object. Given that i cannot easily turn “Q42” into a date, it is difficult to create line graphs to see team improvement (unless im missing something here, please point me in the right direction).

@Bob_The_All_Mighty

and do you think the step up in quality is worth it ?

Yes. While i do love GSheets and the wonderful things i can do there, i do like the simplicity and speed of things i can do with Tableau. There is no longer a need to sit down at my computer for 6 hours trying to figure out one formula to pull the exact data i want from a giant sheet. Tableau does all the tedious work for you, and in a limited amount of time, this is extremely helpful. Updating isnt too bad, and as long as you 1. Upload with “Keep Google Sheets Linked and Embed Username” (or something along those lines) checked, and 2. Know the password to your Tableau Public account, you can easily update your data within minutes straight from Google Sheets, without even needing to go into Tableau Desktop.

Another advantage to Tableau is this right here

You can easily link multiple Google Sheets together and combine the data in order to see everything in one place

TL;DR:

Tableau has more pros than cons, and at this time i definitely recommend Tableau over GSheets for any type of large data analysis.

---

Here is a Team Lookup i made for my team last year using Google Sheets (this shows general data pulls, and graphs at the bottom).

And

Here

is a quick Dashboard (inside of a story) that i made for my team for this year (featuring 5410). This is still in the works, but i was having fun with it

Feel free to ask me any questions. I would be happy to better explain myself, or share data examples i have.

[Zentangle]

I’m going to echo @Kaitlynmm569 here, as they stated all the same points I would have used. If you happen to have any issues at North Star, feel free to stop by our pit or come looking for me in the stands, I’d be glad to help.