Monitoring Study Responses

Accessing the Responses Dashboard

Responses in the sidebar (/responses/).

The page lazy-loads as you scroll. The counters read "Loading…" and the charts stay empty until you scroll them into view — that is not a fault.

Dashboard Overview

Study Information

• Study Name: the name of the current study
• Interviews with Responses: interviews in which the participant answered at least once
• Completion Rate: how many of those interviews were completed

Key Metrics

1. Participation
2. Interviews with Responses — interviews where the participant answered at least once. Someone who opened the link and left without answering is not counted here.

3. Both toggles in Filter Interviews apply to this number.

4. Completion Rate

5. Number of completed interviews
6. Visual progress bar showing completion against target (if set)
7. Percentage of completion displayed

This tile and get_study_stats count different things, on purpose. The page counts interviews that produced at least one answer; the API's total_interviews counts every interview row, including links that were opened and abandoned. The API number is normally the larger of the two, and the gap between them is your drop-off before the first question.

Data Visualizations

1. Response Trends

• Purpose: Track interview activity over time
• Features:
• Daily interview volume
• Completion trends
• Comparison between started and completed interviews
• How to use:
• Hover over data points to see exact numbers
• Use the legend to toggle between different metrics
• Zoom in/out to focus on specific time periods

2. Question Progression (Sankey Diagram)

• Purpose: Visualize how participants move through your survey
• Features:
• Flow of participants between questions
• Drop-off points where participants exit the survey
• Most common paths through the survey
• How to use:
• Hover over nodes to see question details
• Hover over links to see how many participants took that path
• Identify potential issues with question flow or difficulty

Every interview that enters a question leaves it, either onward to the next question or into Abandoned, so the widths across any column add up to the width coming in. The Qn labels match the ones in the duration table below: Q1 is the study's first question in both.

3. Interview Duration Analysis

Summary Statistics

• Median Duration: The middle value of all interview durations
• Mode Duration: The most common interview duration
• Range: The spread between shortest and longest interviews

Per-Question Timing

• Median Time: Typical time spent on each question
• Mode Time: Most common time spent on each question
• Time Range: Minimum and maximum times for each question

Filtering Data

Filter Interviews has two toggles — Only Completed and Only Simulated — applied with Apply Filters. They affect the trend chart, the question progression diagram and the duration statistics.

With MCP

get_study_stats {"study_id": "…"}
list_interviews {"study_id": "…", "completed": false}

get_study_stats returns total, completed and incomplete interviews, the completion rate, and quota fill and remaining when the study has a max_responses quota. Simulated interviews are counted separately. The charts themselves are dashboard-only.

Best Practices for Monitoring

1. Regular Check-ins
2. Monitor response rates daily during active data collection

3. Set up notifications for significant changes in response patterns

4. Identifying Issues

5. Look for sudden drops in completion rates
6. Note questions with unusually long or short response times

7. Monitor for technical issues or confusing questions

8. Data Quality

9. Check for patterns in drop-off points
10. Monitor average completion times for consistency
11. Look for unusual response patterns that might indicate low-quality data

Troubleshooting

No Data Showing

• Ensure the study is active and shared with participants
• Verify that participants have started the survey
• Check your internet connection if data isn't loading

Incomplete Data

• Partial responses won't appear in the "Completed" count
• Check the Response Trends chart for partial completions
• Consider sending reminders to participants who started but didn't finish

Reading a zero completion rate

A study that shows plenty of started interviews and no completed ones usually means people are dropping out — but check the transcripts before you rewrite your questions. Open a few from the Interviews page and see where they stop. If they all stop at the same point, and especially at the same kind of turn, the cause is likelier to be technical than motivational.

Next Steps

1. Read transcripts — Interviews
2. Analyze — Clusters & Charts, Thematic Analysis
3. Recruit more participants — Recruit

Getting Help

For assistance with the Responses dashboard: - Refer to our Help Center - Contact our support team - Check our knowledge base for common issues

Last updated: September 2026