- Evgenia Shestunova
9 Tips to Improve Your Data Visualizations
Having data visualizations that are clear, easy-to-follow, and are visually appealing can make or break any report. At Kai Analytics we are incredibly fortunate to work with Amber and Adriana as they make sure that our graphs and other data visualizations help us tell better stories with our data. We decided to write this blog post because we wanted to share their expertise and their tips for improving data visualizations. We hope you will find these tips useful!
1) Don’t crowd your data visualizations
This one is an easy miss, but this is really a situation where less is more. When you start crowding your data visualizations, it becomes hard to discern what is important and what isn't. Get proactive about using white space and utilize it in a way that will emphasize your result and make your reports more visually appealing.
In addition to making your data visualizations easier to read, using white space will also help you look more polished and professional. A report that uses white space well can help you look calmer (as a brand/individual/institution/etc.) and more confident in your findings.
2) Use visually engaging colours consistently
Pick a few colours (2-3 is usually a good number) and stick to them. Consistency is very important for visual design, especially for data visualizations. Make sure that you are using the same colour for the same type of data as that is key to making your data visualizations look polished, professional, and easy-to-follow.
Also, if you are reporting on the same data category (e.g., employee satisfaction), try to use the same colour in all your graphs throughout your report. Using the same colour will improve comparability of data and your reader will definitely thank you for that.
3) Break up your graphs into facets
If you find yourself in a situation where your graphs are starting to look messy and you have too many data labels to compare, consider breaking up the graphs into facets. Faceting allows you to create a separate graph for each subset of data. When placed side by side and using the same axes, you can quickly compare the similarities and differences across multiple labels.
4) Label your data visualizations
Label your visuals clearly. The reason why it is important to give your visuals labels is because some people that are reviewing your data visualizations in reports or are looking at them on a website may simply be skimming the page. When a person is briefly glancing at a graph, they might not have time to come to a conclusion, but they still want to get as much information out of your graph as possible. Therefore, label your graphs and other data visualizations to make things more digestable for your reader.
5) Label the alt-text for your data visualizations
This tip is not as intuitive as the rest but it is important if you will be posting your data visualizations online. Alt-text is alternative text that is used when an image (e.g., a data visualization) is not loading or when a user chose to not view images. An example of a situation when alt-text is used is if someone that is visually impaired is listening to your report or webpage with a screen reader, the screen reader will read out that alt-text for accessibility purposes. Also, alt-text will be “read” by search engines (e.g., Google) so it can learn what your webpage or report is about and rank your web page.
All in all, fill in the alt-text for your data visualizations and if you are not sure how to explain what your data visualizations are about, check out this very straightforward guide here.
6) Use the right type of chart to communicate ideas
You might have found the most compelling data, but if you are not presenting it in an appropriate way then your findings risk being overlooked. The rule of thumb is:
Comparisons --> bar charts
Trends --> line charts
Relationships and distributions --> scatter plot charts
Simple compositions --> pie charts
There is also a fantastic selection diagram (not created by us and we are also not paid to promote it in any way) that can help you identify which graph you need to use for what type of data that you can access here.
7) Make your data visualizations audience specific
Keep in mind who your audience is and ask yourself these questions:
Are these people familiar with the subject matter?
Do these people use particular “language” to talk about this subject?
For example, if you are presenting something very technical to C-suite level executives you should probably visually represent your data and describe it differently than if you were explaining it to a co-worker in the same department.
Also, try to align your language to your audience’s priorities. For example, C-suite level executives might be deeply invested in cutting costs and increasing profits so if you know you will be presenting your data visualizations to them, make sure your descriptions explain how that data relates to that.
8) Standardize the axes of your graphs
If your data visualizations are graphs that discuss the same data – make sure that the axes are standardized. The reason why we have this tip included is because when you automatically make a graph in Excel, Excel will scale your axes (look at the image above). While this may look more aesthetically pleasing, it makes comparing data harder and can be misleading for your reader. Make sure to manually standardize your axes if you do choose to create your graphs in Excel.
9) Make word cloud with phrases, not with words
If you are creating word clouds, try to create them with phrases and not just words. The reason why this is important is because using phrases gives your data visualization more context. It will help your reader better understand what the word cloud is actually talking about than just randomly floating words.
We hope that you enjoyed this article and found our data visualization tips useful! If you would like to find out how Kai Analytics can help you with your data visualizations, please book your free consultation and we will be happy to help you.