Figma to Dovetail: end-to-end research workflow guide

A Figma-to-Dovetail research workflow runs in five stages: prepare the prototype, recruit external participants, run sessions, export artifacts to Dovetail, and synthesize insights. When each stage is set up correctly, a team can move from share link to repository-ready findings in five to eight business days.

The workflow sounds simple in theory. In practice, most teams lose time at two specific chokepoints: sourcing participants who actually match the target persona, and moving raw session data into Dovetail without losing context. This guide covers both.

Why external participants matter for prototype testing

Internal colleagues and warm contacts carry assumptions that skew prototype feedback. They already know the problem you are solving, they interpret unclear UI more charitably, and they rarely behave like a first-time user. None of that shows up in a usability log.

External participants, ideally screened strangers from your target segment, behave closer to real customers. They get confused where real customers will get confused. They complete tasks at the rate real customers will complete them. That signal is what makes the data worth putting in Dovetail at all.

For a deeper look at sourcing participants specifically, see how to recruit participants for Figma prototype testing.

The five-stage workflow at a glance

Stage	What you produce	Typical time
1. Prepare prototype	Shareable view link, task brief	1 day
2. Recruit participants	Screener, confirmed participants	1 to 3 days
3. Run sessions	Recordings, notes, task logs	1 to 2 days
4. Export to Dovetail	Transcripts, files, highlights	Half a day
5. Tag and synthesize	Insight notes, themes, shared report	1 to 2 days

Stage 1: Prepare your Figma prototype for external testing

Before you send a single share link, lock down the prototype presentation so participants experience what you intend.

Set the share permissions correctly. In Figma, open the prototype, click Share, and set the link to “anyone with the link can view.” Turn off “can inspect” so participants cannot see developer annotations. Prototype links open in the browser without requiring a Figma account, which removes a common drop-off point for external participants.

Write a task brief. External participants need context without contaminating the test. A one-paragraph brief explains the scenario without naming the solution you are testing. For example: “Imagine you are a procurement manager looking to run a vendor evaluation. The prototype you are about to see is a tool that might help. Please complete the tasks below as if this were a real product you are trying for the first time.”

Define measurable success criteria. For each task, decide in advance what counts as completion and what counts as a failure. This consistency matters when you later tag data in Dovetail. Common criteria include: task completed without assistance, number of misclicks before correct path, and subjective confidence rating on a 1 to 5 scale.

For a broader view of prototype testing best practices, see the prototype testing guide: best practices for effective user validation.

Stage 2: Write a screener and recruit external participants

The screener is the single biggest variable in data quality. Over-screening produces too-few qualified participants; under-screening fills your sessions with people who do not match your persona.

Screener length and structure. Keep screeners to eight to ten questions maximum, with three to five qualifying criteria. Start with role or job function, layer on tool familiarity or domain knowledge, then close with availability and consent.

Sample size guidance. For moderated sessions, five participants per distinct user segment is the widely cited threshold from Nielsen Norman Group for surfacing the majority of usability issues. Two distinct segments means ten participants. For unmoderated runs, 15 to 30 participants provide enough data to spot click-path patterns statistically.

Incentives. Consumer participants expect $15 to $25 for a 30-minute unmoderated session. B2B professionals in specialist roles expect $75 to $150 per equivalent hour. Under-incentivizing a B2B screener is the fastest way to erode response quality.

Recruitment channels. Verified research panels are the fastest path to screened strangers. CleverX surfaces matched B2B and B2C participants from a verified panel of 8 million+ across 150+ countries, with most studies fielded within one to three business days. If you prefer manual outreach, targeted posts in relevant Slack communities and Reddit threads can work for consumer audiences over a longer timeline.

Stage 3: Run sessions

You have two broad options: moderated live sessions or unmoderated self-directed runs. Neither is universally better; the right choice depends on what you need to learn.

Moderated sessions work best when you are exploring unfamiliar territory or testing complex multi-step flows. The moderator can probe reasoning, ask follow-up questions, and catch the moments where participants go silent rather than voice confusion. Record with Zoom, Google Meet, or Loom. The best Figma prototype testing tools can also host sessions with built-in screen recording.

Unmoderated sessions work best when tasks are clearly defined, the prototype flow is complete, and you want click-path data at scale. Platforms like Maze generate Figma-compatible share links, record screen and voice, and log click paths automatically. Participants complete the study on their own schedule, which speeds up fieldwork.

What to capture regardless of method:

• Full session recording (video plus audio)
• Screen activity or click-path log
• Post-task rating for each task
• Any verbal comments or written think-aloud notes

Stage 4: Export artifacts and import to Dovetail

This stage trips up many teams because Figma and Dovetail have no native two-way integration. The handoff is manual, but it is faster than it looks once you have a standard process.

From moderated sessions. Export Zoom recordings as MP4 files. Upload each recording to a Dovetail project. Dovetail will auto-transcribe the audio, typically within a few minutes per hour of recording. Name the file with the participant ID and session date so you can cross-reference later.

From unmoderated sessions. Export session recordings and any written response data from your testing platform. For summary statistics (task completion rate, misclick counts), paste them into a Dovetail note document as structured data alongside the qualitative observations.

Anchoring to prototype screens. Dovetail does not pull Figma frames automatically. When a finding is tied to a specific screen, paste a screenshot of that frame into the highlight note in Dovetail. This keeps the visual context with the insight rather than buried in a separate file.

Embedding the prototype link. Add your Figma prototype view link as an embedded resource in the Dovetail project. This gives any team member who reviews the insights a way to navigate the prototype directly from the repository.

You can read more about managing Dovetail projects alongside external participant data in Dovetail review 2026.

Stage 5: Tag, theme, and synthesize in Dovetail

With transcripts and recordings in Dovetail, the synthesis work can begin. The goal is to move from raw session data to a shareable insight note that your product team can act on.

Tagging. Work through each transcript and highlight key observations. Apply tags to describe what is happening: navigation confusion, positive reaction, task failure, feature request. Create a shared tag taxonomy before you start so two researchers coding the same transcript land on consistent labels.

Theming. Group related tags into themes. A theme might be “checkout confusion” or “trust signals work.” Dovetail’s insight note view lets you pull highlights from multiple sessions under one theme, which is where patterns across participants become visible.

Sharing findings. Use Dovetail’s share feature to send the project to stakeholders as a read-only link. Include a one-paragraph summary at the top of the project board with the three to five headline findings and the recommended next steps. This format works for async reviews with product managers and engineers who were not in the sessions.

For teams deciding whether Dovetail’s repository capabilities or a more integrated end-to-end platform better fits their workflow, CleverX vs Dovetail: end-to-end vs analysis-only covers that tradeoff directly.

Common pitfalls that break the workflow

Skipping the task brief. External participants given only a share link with no context produce noisy data. They explore freely rather than completing defined tasks, which makes Dovetail tagging inconsistent.

Recruiting from warm contacts. Friends, colleagues, and LinkedIn connections in your network already know your product space. Their feedback patterns differently from genuine first-time users, and you cannot control for that bias in analysis.

Not naming files consistently. When recordings and transcripts lack a common naming convention tied to participant IDs, cross-referencing data in Dovetail becomes slow. Set a naming format before fieldwork starts (for example: P01-2026-07-07-session.mp4).

Tagging without a shared taxonomy. Two researchers tagging the same session without agreed tag definitions produce inconsistent data. Themes built on inconsistent tags lead to misleading conclusions. Align on a core tag list before anyone opens a transcript.

Frequently asked questions

How long does a Figma-to-Dovetail research cycle take end to end?

A standard five-participant moderated prototype study typically takes five to eight business days from prototype ready to insights in Dovetail. Recruiting through a verified panel takes one to three days, sessions run over one to two days, and tagging takes another one to two days. Unmoderated studies with 10 to 20 participants can compress the timeline to three to five days because sessions happen in parallel.

Do I need a Figma paid plan to share prototypes with external participants?

No. Figma free plans allow you to share prototype view links with anyone, including external participants who do not have a Figma account. The participant simply receives a URL and opens the prototype in their browser. You only need a paid plan if you want to unlock advanced prototype features such as variables or if your team needs more than three active projects.

How do I get session recordings into Dovetail from prototype testing?

For moderated sessions recorded in Zoom, Loom, or Google Meet, export the recording as an MP4 and upload it directly to a Dovetail project. Dovetail will auto-transcribe the audio. For unmoderated sessions from tools like Maze or Useberry, export the session recordings and notes, then import them into Dovetail as files or paste text into a note document. Dovetail also has a Zapier integration that can automate transcript imports from certain sources.

What is the ideal number of participants for Figma prototype testing?

Nielsen Norman Group research suggests five participants per distinct user segment surface most critical usability issues in a moderated study. If you have two meaningfully different user types, recruit five per segment for ten total. Unmoderated studies benefit from larger samples of 15 to 30 participants when you need quantitative click-path metrics alongside qualitative observations. Start with five, then scale up if major issues remain unclear.

Can I run this workflow without a dedicated recruitment platform?

Yes, but timelines will extend significantly. LinkedIn outreach, relevant Slack communities, and your own CRM contacts can yield willing participants, though you will spend two to three times more hours on coordination. The main risks are respondent bias from warm contacts and slower turnaround. A verified research panel removes both problems by surfacing pre-screened strangers who meet your criteria, typically within 24 to 48 hours of posting your screener.

Does Dovetail integrate directly with Figma?

Dovetail does not have a native two-way Figma integration. However, you can embed Figma prototype view links inside Dovetail notes as context for your findings, and you can paste frame screenshots from Figma into Dovetail highlight notes when you want to anchor a tag to a specific prototype screen. This manual workflow is sufficient for most teams. A purpose-built usability platform like Maze or Useberry bridges both tools more tightly if you need automatic screen-level data capture.