At Coda, we’re obsessed with data, and we use a number of visualizations to help us understand things like how our community uses new features and how new customers find us. More often than not, we’re viewing charts based on numerical values, but what happens when the data isn’t a raw set of integers, but text-based feedback from our users?
Starting today, you can use Word Clouds as a Chart Type to visualize your data tables, and to understand the frequency of words within your selected data set:
Word clouds are great for writers to understand if they’re “overusing” certain words, and for teams to find key themes in their sentiment tracking or feedback analysis workflows. While testing this feature, one of our own product teams was able to dig into key feedback themes from user research on a totally different product improvement we’re testing (stay tuned!)
You can find this option under + > Chart > Word Cloud.
When configuring your word cloud, you have control over which data sets (columns) contribute text for analysis and visualization. If you want, you can also exclude a set of common English Language words and/or indicate your own set of words for exclusion.
We look forward to hearing about how you use this feature, so please feel free to share your feedback in the comments below!
Don’t misinterpret my next few comments; I love Coda, but… indulge me while I rain on this little parade a bit.
That’s good. And because of this obsession, you should have no difficulty understanding why I’m thrilled and completely annoyed about this new feature.
The vast majority of words in Coda documents cannot be transformed into data for any process or purpose. As such, word clouds are pretty much out of touch with – wait for it – all of the texts in documents.
Even the AI-based sentiment Pack is unable to derive sentiment from document texts or comments.
The spirit of Coda is a complete reimagining of the idea of documents, yet the most important silo of information – text – remains sidelined from all that should be central to business processes, embedded unstructured data, automation, and meaning.
And please don’t interpret this commentary as a feature request to add section texts as a source for word clouds; that’s not the issue. Text and conversations simply need to be exposed as first-class citizens to play right along side data and so they can become data.
Hey Bill, thanks for the feedback. Totally agree that sections and comments are data too. We’ve definitely talked about features that would expose this data to users and allow them to do cool things with it. Would be cool to get someday and it’s on our list
A bit disappointed when I went on a document full of text and when I added a World Cloud it was asking me for a… table… I thought this was targeted to writers. Which writer writes their content on a table?
A word clouds of tags would be very nice way to do some filtering.
Is it possible click on the words in the cloud words to apply the filters?
Also, would be nice to have the table resizable.
)