The Top Free Open-Source Data Visualization Tools in 2022

How many free and open-source visualization tools can you name? I listed more than 40 — an overwhelming number! Given these options, how do you sort through all the options and find the best one for your purpose?

Kind of Free

But wait. Let’s clarify what we mean when we say “free and open-source.” Open-source tools are always free. Free tools are not always open-source. 

Software that is free but not open-source often has a free tier. The free tier has limitations. The free version may be enough. Or you may find that you need to pay for a commercial license to get additional features.

Both Power BI and Tableau are free but not open-source. Both offer a free license with limited functionality. 

The free version of Tableau (Tableau Public) offers limited data connections and will only save work to the Tableau Public website. Tableau Public is good for learning the tool or for doing data journalism. But it won’t work as a solution for most corporate applications. 

The free version of Power BI cannot share content with others, but you could try the Javascript library Highcharts, which offers a free non-commercial license. 

Free and Open-Source Tier

I said earlier that open-source tools are always free, and that’s true, but some software has a proprietary option in addition to its free, open-source software. The proprietary portions offer enterprise customers additional features with the paid version.

For example, Plotly’s Dash offers a free & open-source tier. But as they explain, “when building Dash apps in a business setting, you’ll need Dash Enterprise to deploy and scale them, plus integrate them with IT infrastructures such as authentication and VPC services.”

Similarly, R Shiny has Shiny Server open-source, but also offers commercial licenses. If you need to integrate with existing authentication systems, you might need to spring for a commercial license. A google search did turn up a post explaining how to add authentication to the open-source version of Shiny Server.

Datawrapper has a version that is free for commercial use and can publish privately. It’s not fully open-source, but part of it (Datawrapper Core) is open-source and available on GitHub. The dashboard and maps are gorgeous, and it’s responsive for mobile users. Unlike Tableau and Power BI, the free tier does allow for private sharing of visualizations. So, what’s the catch? Datawrapper doesn’t offer direct connections to databases. It only visualizes data stored in Excel, Google Sheets, and text files, so you’d need to transform your database tables into one of those formats before creating visualizations. Also, larger organizations may want to invest in the enterprise license for added security options. 

Fully Free and Open-Source

Then there are the completely free and open-source projects. Many of these are modules or packages for open-source programming languages:

For Python, Matplotlib is the most popular, but there are others, including seaborn, Altair (for statistics), and Bokeh (for interactivity). 

For R, ggplot2 is the most popular. Lattice (for multivariate) is also popular. You can also create graphs with base R. As with the Python libraries mentioned, these work well for data exploration.

For JavasScript, there is D3.js, Plotly, Chart.js, Recharts, Victory, ReactViz. These libraries are suited to creating interactive charts on the web.

Some of these work in multiple languages. Plotly is an open-source Javascript library, but Plotly for R lets you use it with R, and the Plotly Python library lets you use it in Python.

Free Tableau / Power BI Alternatives

There are numerous open-source data visualization tools, but many of them require developers to write code. This can slow development time and make it more challenging to find analysts with the requisite skills.

But what if you need a visualization tool that doesn’t require you to write code. And you don’t just want a project with a free tier, but a product that is full-featured for commercial use. And you need to connect to a wide variety of data sources, including files and popular databases. If you plan to use it to create internal corporate dashboards and reports, you need integration with your company’s existing authentication. You likely also want the ability to choose between managed, on-prem, and cloud hosting.

In other words, are there any fully free alternatives to Tableau and Power BI? Not just open-source libraries, but low or no-code data viz applications designed for the enterprise.

Yes — Redash, Metabase, and Apache Superset. However, they don’t appear to have a significant market share compared to Tableau or Power BI. A search on Indeed produces over 5,000 job listings mentioning Power BI but only around 100 for Superset & Metabase and 50 for ReDase. 

Niche-Specific Visualization Tools

The tools mentioned so far are general-purpose tools that can take tabular data and create bar, line, and pie charts. But if you’re looking specifically for interactive mapping, network visualization, or system monitoring you might consider the following niche tools.

• Interactive mapping – Want a great-looking interactive map on your webpage and don’t want to pay a licensing fee? Try Leaflet, OpenLayers or Polymaps.
• Network analysis – Need to visualize who’s retweeted who, who’s friends with who on Facebook, which researchers have published papers together, or which websites link to which websites?. Most standard BI tools cannot visualize the network, but network-specific tools like Gephi can create visualizations like this one that look at relationships between the characters from Les Miserables.

• Monitoring System Logs – Another important niche is visualization tools focused on monitoring time-series data typical of IT system logs. Basically the server version of what you see on the Performance tab of the Windows task manager. Kibana is a popular AWS tool for visualizing log data and is tightly integrated with ElasticSearch. Grafana is also popular.

Market Demand for Data Visualization Tools

It’s great to talk about features, but how much market share do these different options actually have? Are companies using the open-source options or paying for the commercial options?

To get a sense, I did searches on Indeed.com for the different categories. Here’s what I found:

Searches like this aren’t perfect. A few tools, like Mode, had names that are also common words used in job descriptions. There were over 54,000 posts for the “mode” keyword, but I think the vast majority aren’t related to the software.

But from what we can see, open-source has only a toehold in the market. Tableau and Power BI are the 300-pound gorillas. However, a search for Python positions including the term “visualization” results in more than 12,000 posts, so you could say that there are a lot of python data analyst/data scientist roles that involve some degree of visualization skills. Also, I did find around 2,000 postings for Kibana and Grafana but left them off the bar chart since they’re focused on system monitoring rather than general visualization.

So, What’s the Best Free or Open-Source Tool?

Oh boy! It depends on your use case. Let’s exclude the geographic mapping, network, and log monitoring tools and exclude Power BI, Tableau, Dash, and Highcharts for enterprise because their free tier is missing many necessary features. The open-source version of R Shiny also does not scale well to the enterprise, though there are workarounds. 

Here are some starting points:

• If you’re looking to learn a viz tool to make yourself more marketable, start with the free versions of Tableau and Power BI. 
• If you’re trying to create visualizations for data exploration, Python and R have numerous excellent open-source options. 
• D3 leads in terms of general-purpose Javascript visualization.
• If you want to create good-looking graphs from an Excel or CSV file, try DataWrapper. I used it to create the bar charts you see above!
• For a fully open-source option with a similar feature set to Tableau and Power BI, try ReDash (free when hosted on your own server), Metabase, or Superset.