How we built AI in Dovetail—magic analysis

Analyzing customer data can be a daunting task. Whether you’re digging into dozens of interview transcripts or synthesizing survey results, the process is often repetitive and time-consuming. That’s why we created magic analysis in Dovetail—to take the heavy lifting off your plate and let you focus on the insights that matter.

Today we’re delving into our approach to AI integration in magic analysis and how you can harness Dovetail’s AI capabilities effectively to get closer to your customers. Uncover patterns, identify themes, and make sense of complex data—all while staying in control of your workflow.

Why use Dovetail’s magic analysis?

Dovetail’s magic analysis was designed with specific challenges in mind:

• For researchers, who need to process dozens of customer interviews, our AI cuts through the noise to surface key themes and insights.

• For designers, who want actionable takeaways from usability tests or surveys, magic analysis helps pinpoint what matters most without you having to wade through raw data.

Imagine you’re juggling fifteen customer interview transcripts while trying to extract recurring themes. It’s a race against time to share insights with your team before the next feature sprint begins. Or perhaps you’ve just wrapped up a usability test. Your goal is to summarize key takeaways, but sifting through hours of sessions requires more time than you have.

From tagging key highlights in transcripts to summarizing survey responses and uncovering overarching themes, magic analysis removes the manual grind so you can focus on making informed decisions. It does this by harnessing the power of AI to make sense of your customer data instantly. 

With magic analysis you can:

• Automatically find and highlight key moments in your data

• Cluster highlights into groups on your canvas

• Summarize key points in notes and insights and more 

It takes the grunt work off your plate so you can focus on what matters—all while keeping you in the driver’s seat.

How Dovetail’s AI works

Our AI approach is guided by three core principles to ensure that magic analysis fits seamlessly into your workflow while keeping you in control:

1. Human-in-the-Loop: with Dovetail, you’re always in control. You can toggle between automation modes—off, suggest, or on—depending on your preferences and comfort level. Whether you want full automation or occasional nudges, the choice is yours.

2. Trust and transparency: we believe you should always know what’s happening with your data. AI-generated contributions are clearly marked with the magic contributor avatar, and edits or reviews are attributed to team members. This transparency ensures you can trust the outputs and understand their origin.

3. Aligned with your workflow: magic analysis doesn’t replace your processes; it enhances them. For instance, features like magic highlighting replicate manual steps like tagging and categorizing but streamline them significantly. This alignment makes it easy to see what the AI is doing and make adjustments as needed.

Best practices for using magic analysis

To get the most out of magic analysis, follow these tips:

Use predefined tags

Start your project with templates or tag packs designed for specific use cases, such as usability testing or customer interviews. These tags are optimized for Dovetail’s AI, ensuring accurate results without extra setup.

Provide examples

Manually tagging a few examples gives the AI a reference point to learn from. By showing the system what you’re looking for, you can improve the quality and relevance of its output.

Write clear tag descriptions

AI works best when it understands your goals. Writing a tag description, rather than one to two words in your tag gives the LLM extra context on what to look for. 

A good tag description will vary depending on the tag, but it can be helpful to think of your tag descriptions as instructions for a new coworker—simple, clear, and to the point. For example: instead of using vague tags like “Positive,” provide a concise description, such as: “Mentions of positive feedback or successful outcomes.” 

Simplify your workflow today

Dovetail’s magic analysis is here to help you work smarter, not harder. By automating repetitive tasks like tagging and summarizing, it frees you up to focus on deeper customer understanding.

Ready to put magic analysis to work? Try it in your next project or explore our help article for more tips and tricks to make the most of Dovetail’s AI capabilities.