Tips for making a successful Original Content [OC] post
The following are tips collected by successful DataIsBeautiful contributors. Consider them useful guidelines for making a successful post to DataIsBeautiful.
Anyone with 50 karma in this subreddit can edit this page.
1. Read the rules in the sidebar
Your post will just get removed if you don't follow the rules of the subreddit. Read the rules carefully.
2. Labels and Legends
What does blue mean? What's the vertical axis? Make sure everything is labeled.
Make sure that the font sizes are big enough to be readable.
Never save as JPEG. It ruins text. Always use PNG.
5% of the population has some form of colorblindness. Try to avoid green-red color schemes because that's the most difficult for many color blind people.
The defaults in Tableau and GGplot have been vetted. Other color options can be found at ColorBrewer.
You can simulate what your visualization will look like to a colorblind person using this tool. It's recommended to run your color scheme through the tool before committing to it.
5. Data and Code
Including the data and code behind your visualization allows others to learn from and analyze your visualization.
Explain where the data came from, and post a link if possible.
If you collected the data yourself, please consider posting it. For text (such as CSV), try Gist.
Give the submission an informative title that isn't too editorialized.
Include at least a short description of how you acquired, manipulated, and visualized the data in the comments.
The best time to post is generally between 8am and 12pm EDT (UTC-4). Other times also work well, but most of the successful posts on DataIsBeautiful were posted in that time range.
Consider submitting your work to the Wikimedia Commons for use in Wikipedia as well.
7. Misc Tips
Dropping grid lines can help clean up a complicated plot.
Avoid legends. Put the labels directly inside the image, like this.
Consider putting the data values directly in the visualization, like this.
revision by rhieverViz Practitioner | Randy Olson— view source