Create and maintain tags

Tags are an important part of performing analysis in Dovetail. They help you connect the dots between important moments surfaced as highlights, either within a single piece of data (like an interview) or across multiple pieces of data (multiple interviews) that make up a whole project.

Tags thematically group bite-size information captured within your data quickly and can be titled anything that captures the theme of the highlighted text. In this lesson, learn ways to use tags to analyze data in projects and make the most of magic highlight with effective tag descriptions.

By default, tags are local to a project, meaning those tags will only appear as options for the single project they are connected to. To effectively manage and group your tags, you can set up tag boards on a project level.

Anyone with Full or Edit access to a project will have the ability to create and manage the tag board for that project.

Bottom-up tagging

You can create new tags organically when highlighting notes in a project. From there, you can view and organize these tags into groups on a board under Tags.

On this board, you can define groups and color-code related tags. You can also move tags across groups, merge related tags together, and cluster your observations into themes that can be elevated and talked to within an insight.

Top-down tagging

If you are looking to apply a top-down approach to tagging, you can also create individual tags, install a community tag board to customize or import existing tags on a board before highlighting your notes. This is great for standardizing your tags for any kind of research including usability testing.

You can improve the quality of our magic highlight feature by adding descriptions to your tags. What makes a "good" tag description will vary depending on the tag and what you are looking for but there are, however, some general guidelines and recommendations you can follow –

• Use the bare minimum required words

  • If you can say it in three words, use three words instead of a sentence. A guideline is to try and not use any more than 50 words**

• Repeat the same word

  • Do not use different words for the same meaning, even if they are synonyms. Repeat the same word.

• Be complete with your information and explanation

  • If a new intern wouldn’t understand the term, assume AI wouldn’t either.

• Use examples if necessary – If you are providing examples, start with the things to consider and follow it with examples.

  • With this, it is better to provide examples of what to do/what to know rather than examples of what

    not to do. What not to do is not a good thing for an LLM unless absolutely necessary.

  • Do not make an example a massive chunk of text. Use something concise, or bullet points, if you’re going to do it. A bad example can throw the LLM off track.