QuickSight Visuals - Bar Charts

Hello everyone, my name is Duncan. I'm a data engineer at West Loop Strategy, an AWS partner. For our QuickSight Visual Series, I'm going to be talking about bar charts.

Bar charts are a really common visualization in dashboards to graphically display data by groups. Now, I personally find them very useful for comparisons of data, showing change over time, and communicating growth for our clients. QuickSight comes with a really robust range of options and styles for your bar chart so that you can really customize and experiment with what works best for your Dataset. The differences in styles include orientation, as well as if they are clustered bars or they're stacked on top of each other either by value or by percentage.

For today, I'm going to do something simple. I'm going to choose the vertical bar chart. For the field wells we have X axis, value, group and color by, and then small multiples. For the X axis I typically just choose dates, but technically, you could use some of the same field wells that you would add to your group and color by, it really just depends on what makes sense for your data. So, I'm going to add order date. This is a date-time field, and I'm going to change it to this to year in my aggregate. I'm going add sales to my value and I'm leave this as a sum. Then, let's say I wanted to display my growth in sales by year, and by the business segment. So, Enterprise strategic or SMB. I'm going to add segment to my group and color by. So without doing too much, we can see that I can show my boss or my client their growth in sales by business segment year over year.

Let's say I wanted to add another level or they wanted this for all the regions in their business, but they didn't want to change the general premise or the general basic layout of how the visual looked. That might be a good excuse for me to use that small multiple. So what I'm going to do is add region into the small multiples. Now I have that same chart or my same bar chart but repeated three times from my different regions. Now, I can show the growth in sales by business segment per region and it's not all clustered together and hard to read. They've been separated and it's a great way to be able to add a bit more of a visual element and a bit more gusto to your dashboard. I'm going to remove regions for now just to keep things simple, but in the future, let yourself experiment a bit.

Another thing that I really like about bar charts, and what I really like about QuickSight in general, is the styling for this is very similar to the other visuals. So in the top right hand corner of the visual panel, I'm going to hit the format visual pencil. Now I have my format visual menu open and you can see I can update the title. I can change the X and the Y axis. We'll get back to that. We can update the group by and color to determine how many bars we want. We can remove the legend and we can add data labels and update the tooltip.

So, why is it so important to be able to update the titles of your X axis and your Y axis? I think that oftentimes we get pretty stuck into our own titling in our Datasets. So right now, let's say order date was pretty complex. It really only reflected something that me as the data engineer understood about our data. I could go in here and change this to year or I could change this into sale made, whatever makes sense internally for my team or my client. And now this makes a little bit more sense to them specifically. So it allows you a lot of flexibility to make this more accessible for your user. I also really recommend personally getting rid of the data zoom and the sort options cleans up the visual a bit more. I find the data zoom to be pretty cumbersome.

The last thing I really want to touch on here is going to be reference lines. These are a great way to be able to just do quick things and be able to show quick reference and context for the data. You can set this up in a few different ways, but today, I'm going to use a calculated line. To get that set up I want to use my sales field and I'm going to show average sum of sales. Now, we can see on our chart, that the line appears. What we can do from here is update where the label is, we can change if the line is solid or dashed, and we can also change what side that the line label appears on, and what it says. I highly suggest taking some time to experiment and play around with these reference lines. They're a really great way to be able to give context to your data.

Thanks for joining me today. To learn more about how we help our clients use AWS Cloud technology or if you want to schedule your own QuickSight training session, check out our website or follow us on LinkedIn.