Free trial
The first three interviews are free, with no credit card, demo call or minimum spend required.
Learn moreUser Evaluation
AI-moderated customer interviews where every claim links to its clip
Beginner API
Own User Evaluation? Verify ownership to get a dofollow link
What is User Evaluation?
User Evaluation is a navy eval writer tool. User Evaluation pairs real, screened participants with an AI voice moderator, then delivers a study report within 48 hours in which every claim links to the clip that supports it. A companion Swarm mode registers AI agents as product users so replays can surface friction before you spend on human interviews.
User Evaluation runs voice interviews between real, screened people and an AI moderator, then packages what they said into a single study report where each claim carries a link to the clip it came from. The company's own rule is blunt: if it cannot show a clip, it does not make the claim. Participants are paid real people, not simulated personas, and consent to recording is collected before each call begins. The workflow starts with a typed question. An assistant named Eva drafts a five-question guide plus a screener, and nothing launches until the researcher approves it. Interviews happen as live voice calls in the participant's language (40 are supported), and the moderator follows up on answers in the moment. Researchers can listen in live and swap the next question. Reports arrive within 48 hours with themes, a confidence level per theme, and suggested next tests, exportable to PPTX, DOCX, PDF, Markdown or CSV. A second product, Swarm, registers AI agents as users of a live product and keeps them returning over weeks, recording replays with red marks where an agent hit friction. All material, including agent sessions, uploaded calls, PDFs and CSVs, lands in one shared evidence base that the Analysis layer cites. Participant sources carry fixed per-interview prices: your own users, an everyday panel, or professionals screened by role and seniority. Product teams, founders and UX researchers are the obvious audience, particularly small teams that need directional answers fast and cannot wait weeks for a traditional study. The site itself is careful that three interviews reveal whether you are asking the right question, not reliable percentages. Agencies running frequent discovery will notice the per-interview billing adds up, and anyone who needs statistical rigor should treat the output as a starting point rather than a survey replacement.
How do you use User Evaluation?
1Sign up and open a workspace
Create an account with email or Google. The first three interviews are free, with no card or demo call required.
2Write the question you want answered
Type a single research question, such as why trial users leave in week two. Eva drafts a five-question guide and a participant screener from it.
3Approve the guide and pick participants
Edit the drafted questions, then choose your own users via a link, the everyday panel, or role-screened professionals. The per-interview price includes recruitment.
4Watch or steer the live calls
Each interview is a live voice call with the AI moderator, which asks follow-ups on every answer. You can listen in and change the next question while it runs.5Add agent sessions if the product matters
In Swarm, agents register as users, return over weeks and generate session replays with red marks where they hit problems. Send key findings to real people to confirm.
6Read and share the study report
Within 48 hours you get recordings, transcripts and a report where each claim plays its clip. Export to PPTX, DOCX, PDF or Markdown, or share a view-only link.
Pros and cons
Pros
- Every claim in the report links to a timestamped clip, so findings can be checked rather than trustedAI
- Real screened participants on live voice calls, with recruitment included in the stated priceAI
- Eva drafts the guide and screener quickly, and no interview runs without researcher approvalAI
- Interviews, agent sessions and uploaded recordings share one evidence base and citation formatAI
- Low-quality interviews can be flagged for a free participant replacementAI
Cons
- Three interviews show whether your question is right, not percentages, so small samples cannot support quantitative decisionsAI
- Per-interview pricing ($10, $39 or $99 by participant source) scales steeply for ongoing research programsAI
- Panel quality depends on screening, and the page offers no detail on panel size or composition by marketAI
- Swarm agents are synthetic users, so their findings still need confirmation from real people before actingAI
- Transparency on data retention periods and consent wording beyond the recording agreement is thinAI
How much does User Evaluation cost?
Pricing
Interviews start at $10 with your own participants, $39 with the screened everyday panel, and $99 with role and seniority screened professionals. Each price includes recruitment and every plan includes the full feature set. You pay per interview, and costly actions show a total for approval first.
Learn moreIntegrations
An API and webhooks handle study creation and findings delivery, and an MCP server connects Claude Code or Cursor. No named third-party app integrations are listed on the homepage.
Features
AI moderator that follows up on each answer, a screened paid panel or your own users by link, 40 languages for interviews and transcripts, speaker-labeled time-coded transcripts, clip citations on every claim, PPTX, DOCX, PDF, Markdown and CSV export, free view-only share links, agent session replays via Swarm, approval before any cost, uploads of calls, PDFs and CSVs, plus API, webhooks and an MCP server.
Learn moreFrequently asked questions about User Evaluation
How much does User Evaluation cost?
Interviews start at $10 with your own participants, $39 with the screened everyday panel, and $99 with role and seniority screened professionals. Each price includes recruitment and every plan includes the full feature set. You pay per interview, and costly actions show a total for approval first.Does User Evaluation offer a free trial?
The first three interviews are free, with no credit card, demo call or minimum spend required.Does User Evaluation have an API?
Yes, User Evaluation offers an API.How do you use User Evaluation?
The walkthrough on this page covers 6 steps: 1. Sign up and open a workspace 2. Write the question you want answered 3. Approve the guide and pick participants 4. Watch or steer the live calls 5. Add agent sessions if the product matters 6. Read and share the study report.What platforms does User Evaluation support?
User Evaluation is available on Web App.What does User Evaluation integrate with?
An API and webhooks handle study creation and findings delivery, and an MCP server connects Claude Code or Cursor. No named third-party app integrations are listed on the homepage.
Status
StatusActive
Views0
Outbound clicks0
Added10/21/2026
Platforms
Web App
Pricing
FreemiumPaid
