Effective data visualisation is essential for making your reports as easy to understand as possible. In this article, I’m going to be sharing with you the 7 most important things to consider and implement when designing dashboards. My top tips if you will. Let’s dive straight in!

1. Plan ahead

Before you start adding charts to a report that you’re planning on publishing, make sure that you have a clear idea of its contents and layout.

What I recommend the first thing you do is to write down all of the queries, or questions you want to ask of your data. Make a list. It should contain the name of the query, the data source it comes from (if there’s more than one), the fields in that data needed and the visualisation type.

This last one isn’t essential at this point though if you haven’t done any data discovery to see the kinds of results you’ll be working with. I have my own template that I use for this list of queries. It’s what I share with my clients to help them plan out their project. You can get your hands on a copy here.

Once you’ve made your list, then I recommend taking a blank sheet of paper and a pencil and simply sketching out a rough drawing of how you would like your report to look. Nothing detailed, just a very rough sketch.

Try and add in all of the queries from your list. Believe me this will be a great help. You’ll also find out whether you’re going to need to have more than 1 page in your report, how you could potentially group certain queries together. Which brings me on to my next tip.

2. Tell a story

What I mean by this is that your report is going to be “read” by its viewers, most likely from top left to bottom right, like a page of a book.

That’s just the way our brains are set up. So make sure there is some coherence to the report in terms of which queries you’re showing where. Also, think about grouping queries together that contain data relating to the same area or activity.

What you’ll find with lots of reports that have multiple pages in them is that the first page will contain a kind of summary of the main headlines and the other pages will be dedicated to specific areas and go into more depth. So this is what I mean by telling a story.

3. Avoid TMI

Tip number 3 is closely related to the first two and that’s to Avoid TMI, or “Too Much Information”. What I mean by this is to not overload your report with charts and information making it difficult to read and understand.

You should never be asking yourself “can I squeeze another couple of charts on the page just to avoid going over onto a new page?” Because usually what happens in these cases is that you either end up reducing the size of other charts, making them less easy to read or you move charts around on the page to try and accommodate more charts, which can mess with the story you’re trying to tell.

Or your charts will simply be too close together. And your data needs room to breathe. Rather than overloading your viewer with information, you could even do the opposite, be more minimalist and use as many pages as you need to.

TMI also relates to individual visualisations. For example, don’t have a bar chart with so many bars in it that it’s hard to read and you can’t see the labels properly. There’s a reason why Data Studio by default limits these charts to the top 10 values.

4. Ink to Data Ratio

This is a concept introduced by Edward Tufte, a data visualisation expert, in his 1983 book,The Visual Display of Quantitative Data.

In it he says, “A large share of ink on a graphic should present data-information, the ink changing as the data changes. Data-ink is the non-erasable core of a graphic, the non-redundant ink arranged in response to variation in the numbers represented.”

So, essentially, what he’s saying is that the majority of the ink needed to display (or print) a visualisation should be data ink. For example, the bars on a bar chart or the line on a time series chart.

Wherever possible you should remove anything non-essential to the visualisation. This means things like the grid on a chart, axis labels etc. If the visualisation data can be read and understood without it, then remove it.

A good example would be in a time series chart. The values on the x-axis are dates and obvious to the viewer. So you don’t need to have an x-axis title label that says “Date” because it’s redundant.

An exception to this might be when you have more than one date in the data set and that needs specifying but you get the idea.

If you’re interested in learning more about data visualisation theory then I definitely would recommend you getting yourself a copy of The Visual Display of Quantitative Information. You can find it on Amazon here: https://amzn.to/3jMaGAi

This content is taken from my online course The Ultimate Guide to Google Data Studio. To find out more just head on over to https://www.vitaminbi.education/ to check out a free preview of the course.

5. Choose the right visualisation type

It’s important to make sure you’re choosing the right visualisation type for your query you’re creating. Here are some questions you might want to ask.

Are you comparing values? Then perhaps a column or bar chart would be best.

Are you trying to visualise relationships or hierarchies? Then maybe a treemap.

Are you showing percentages of a total? A pie chart would probably be best here unless you’ve got more than 5 values to display.

Are you working with dates? Then a time series or area chart is the way to go.

All of these questions will help you decide which visualisation type best suits what you’re trying to convey. The goal is to make your data as easy to understand for the viewer as possible and visualisation type plays a big role in achieving this.

6. Use colours wisely

What I mean by this is don’t go wild and start choosing a different colour for each visualisation. Keep your use of colour as simple as possible.

You can use different colours to represent different areas in your report, so everything relating to sales in one colour, marketing another etc. But only when it makes sense to do so.

I guess a rule would be that if you’re going to use a different colour make sure that it’s applied to more than one visualisation in the report. However, you will notice that the best designed reports and dashboards keep the colour scheme very simple.

7. Design for your audience

This relates to not only who is going to be viewing your report, but also how it’s going to be viewed or consumed.

On this first question, “who?”, are they going to need to drill down into the data or filter it? If they are, you’ll need to include these options when building your visualisations and designing your reports.

How familiar are your dashboard viewers with the data that’s being presented? This will determine how much description or non-essential information you have in visualisations to explain what is being presented.

How the report is going to be viewed and consumed will have an impact on things like the orientation and size of your report. Will it mainly be shared as a pdf that will be printed? If so, you’ll need to make sure that it prints properly.

You would also, in this case, perhaps need to add more information like displaying certain values in charts because you can’t hover over a chart with your mouse to display values on a pdf.

So you can see that there are many things to consider when it comes to your audience.

So there you have my top 7 tips for better dashboard design. You can perhaps now understand that it isn’t just a question of throwing charts onto a page and why I think it needs to be studied independently from the other areas of connecting to data and building visualisations, although this last area is somewhat closely related.

This content is taken from my online course The Ultimate Guide to Google Data Studio. To find out more just head on over to https://www.vitaminbi.education/ to check out a free preview of the course.

Get your hands on a copy of my project specifications template here.