If by graphs you mean charts, these are the tools I use: - d3js + nvd3 [0] - if the charts have to be online for consumption of people outside the company. Not

If by graphs you mean charts, these are the tools I use: - d3js + nvd3 [0] - if the charts have to be online for consumption of people outside the company. Not so much a fan of using this because I have to do extra work to convert the data into json, and then write the scripts to generate the chart. - ggplot2 [1] - if the charts are static and need to somewhat be shown (internal analysis, internal reports, client-facing presentation, etc). Of late, I've been using ggplot2 for xkcd-style charts and nobody seem to mind :P - so I don't think people are actually looking closely into the charts - plot (from R) - if the chart is only for my consumption - I don't need a fancy beautiful chart to read a PCA output for example. As for high quality, the charts are only as high quality as the way you use them. If you use the wrong charts for the wrong data, it's not going to be high quality. [0] - http://nvd3.org/‎ http://nvd3.org/‎ [1] - http://ggplot2.org http://ggplot2.org