Dial testing is a powerful tool in market research, particularly when assessing emotional engagement and sentiment in response to video content. Here's a detailed guide on implementing dial testing with the MX8 Labs Research Platform.
Why Use Dial Testing?
Dial testing allows respondents to provide real-time feedback as they watch a video. This method is invaluable for capturing moment-by-moment reactions, helping researchers pinpoint specific parts of the video that resonate or fall flat.
What you can learn from dial testing
Dial testing helps uncover:
- Emotional Peaks and Valleys: Identify moments of high engagement or emotional impact.
- Clarity Issues: Spot sections where confusion or disengagement occurs.
- Audience Preferences: Understand which elements (e.g., visuals, messaging, tone) resonate most.
- Comparative Feedback: Test multiple versions of a video to see which performs better.
How to Code Dial Testing in MX8 Labs
MX8 Labs makes it simple to integrate dial testing into your surveys. Here's how to set it up:
- Play the Video: Use the play_video function to show your video to respondents.
- Add a Dial: Set the dial parameter to position the dial on the screen. Options are:
- "right": Places the dial to the right of the video.
- "bottom": Positions the dial below the video.
- "none": No dial is displayed.
Customize Labels and Captions:
- Use dial_labels to define the scale (e.g., {0: "Dislike", 50: "Neutral", 100: "Like"}).
- Add a custom dial_caption to instruct respondents (e.g., "Adjust the slider to indicate your feelings.").
Example Code
Upload the videos as a material named Creatives. The segment examples use a file named ad-clip.mp4.
from survey import Survey
s = Survey(**globals())
for video in s.get_videos("Creatives"):
s.play_video(
video=video,
start_message="Hit play when you're ready to watch the video.",
end_message="Thank you for watching the video!",
dial="bottom",
dial_labels={0: "Dislike", 50: "Neutral", 100: "Like"},
dial_caption="Adjust the slider to reflect your feelings as you watch the video."
)
s.complete()
Dial-test a video segment
Use start_seconds and end_seconds on play_video to collect reactions to a selected part of a longer video:
video = s.get_video("Creatives", file_name="ad-clip.mp4")
result = s.play_video(
video=video,
start_seconds=10.5,
end_seconds=25,
dial="bottom",
)
Choose a range within the video's duration, with the start before the end. Leaving out the start begins at zero; leaving out the end continues to the source end. Leave out both to play the full video. The same range controls work for secure and standard playback.
Reaction sections describe only the viewed portion, using timestamps from the original video. A section beginning at 12 seconds in this example refers to source time 12, not 12 seconds after the segment started. Pass section.start_seconds directly to get_video_thumbnail; do not add the playback start again. result.total_duration is the source-video playback endpoint, not the segment length.
Use the follow-up loop below with the segment result to ask about its highest or lowest reaction. start_position controls the dial's initial value and does not set where video playback starts.
Ask about the respondent's reaction sections
Use each respondent's dial trace to select moments for immediate follow-up questions. Save the returned play_video result, inspect its sections, and ask about the highest or lowest reaction, a rise or fall, a plateau, or a turning point.
Each section has fractional start and end times, a dial value in your authored scale, one classification, and independent highest/lowest flags. The sections are chronological and do not overlap. Five-second analysis buckets and a flat tolerance of 0.05 are the defaults; bucket_seconds and flat_tolerance let you adjust the analysis.
video = s.get_video("Creatives", file_name="ad-clip.mp4")
result = s.play_video(video=video, dial="bottom")
for section in result.sections:
if section.highest or section.lowest:
frame = s.get_video_thumbnail(video, section.start_seconds)
with s.tag(section.tags):
s.text_question("What were you thinking at this moment?", image=frame)
section.tags retains the section interval, dial value, classification, highest/lowest flags and originating video question on the follow-up answer. See Report on reaction sections for the tag names and values used to compare follow-up answers in reporting.

The frame comes from the whole second containing section.start_seconds. These sections describe the individual respondent's reaction, while the Video Overlay report below summarizes the selected respondent base. Videos without a dial have no reaction sections.
See Media Functions for thumbnail handling and the API reference for the full field and analysis contract.
How Dial Testing Appears in the Survey
Respondents will see the video player with a slider (dial) to the right or below the video. They can adjust the slider in real time to reflect their feelings as the video plays.

How the Reporting Looks
Once data is collected, the reporting dashboard will display:
-
A timeline graph showing the average dial position over time.
-
Key moments (timestamps) of peak engagement or disengagement.
-
Breakdown of sentiment trends by demographics or segments.

For Non-secure media, request a review MP4 with the results shown beside the video using the overlay download workflow. Secure videos are not eligible for these exports.
Report on reaction sections
The reporting tag selector lists tag names. Choose section to report on reaction classifications. Its values are rising, falling, flat, peak and trough; these values do not appear as separate tag names.

In a cross-tab, select section under Questions to show reaction types as rows, or under Cut by to compare them as columns. To filter by a reaction type, select section under Filter by, then choose a value such as falling in the report. See Reporting with MX8 Labs for the report controls.
| Tag name | Reported values or meaning |
|---|---|
section | One of rising, falling, flat, peak or trough. |
highest | True for the section containing that respondent's highest dial rating; otherwise False. |
lowest | True for the section containing that respondent's lowest dial rating; otherwise False. |
start_seconds, end_seconds | The section's inclusive start and exclusive end, in seconds from the start of the video. |
value | The section's dial value attached to a tagged follow-up answer, in the authored scale. |
question_id | The originating video question, linking its sections and tagged follow-up answers. |
highest and lowest are independent of section. For example, a section can have section = flat and highest = True. Use highest = True for the overall highest-rated moment; use section = peak for local peaks.
Tag follow-up questions with section.tags, as in the example above, to retain this context on their answers. Follow-up reports include only the classifications you asked respondents about. If you only ask about highest/lowest sections, you will not necessarily have follow-up answers for every reaction type, even when the video's saved sections include rises and falls.
Thumbnail follow-ups also carry file_name and offset_seconds to identify the video frame shown. The frame offset is separate from the section's start and end times.
Creating a Video Overlay Report
On the Reports tab, select Create → Video Overlay, enter the report name and respondent source settings, then Save to open the editor. You can also set the combined-chart and overlay-download choices during creation. To change a saved report, select it and choose Edit.
The editor has four steps: Scope → Questions & videos → Relabelling → Significance.
- In Scope, set respondent statuses and the optional report window. Project reports also select surveys and whether respondent IDs are consistent across them. Enter dates in the local timezone shown beneath the fields.
- In Questions & videos, choose the required Time-series question. It must provide video filename and time-offset tags (
file_nameandoffset_seconds). The selector only offers eligible questions. - Optionally select one different question under Cut by, such as ethnicity, to group the series. This additional selection is a categorical grouping of the primary measure, not a second time-series measure.

Under Videos:
- Show all videos on one chart combines videos for comparison. When off, each video has a separate chart.
- Generate overlay video enables review-overlay downloads for eligible Non-secure media; it does not start rendering when the report is saved or refreshed. It requires separate charts. Enabling the combined-chart option turns this choice off and makes it unavailable.
- With overlay downloads enabled, Overlay description and Copyright notice set the text used in the export. Use Logo to choose an account image or the default logo. The source video must be Non-secure.
- Excluded tags removes unwanted grouping tags. The filename and time-offset tags are required and cannot be excluded.
In Relabelling, edit the labels for tags and selected response values shown on the chart and generated video. For example, give a grouping variable a clearer name or relabel its categories.

In Significance, set the test, confidence level, minimum respondents, decimal places and maximum filter values. Unweighted uses raw respondent counts instead of survey weights. Save the finished configuration.
Choose an overlay logo
Custom overlay logos come from image Resources in the report's account. Upload the image there before editing the report.
- Open the Video Overlay report and select Edit → Questions & videos.
- Leave Show all videos on one chart off and enable Generate overlay video.
- In Logo, select an account image and check its preview. To use the default logo, clear the selection with the picker's X.
- Save the report. The selection is used the next time you request a review-overlay MP4.
The same Logo picker is available when creating a report. If a saved logo refers to an older survey-media file or an unavailable account image, choose a current account image or clear the selection before saving.
Download a review overlay
Wait for the report's data to be ready, then use the video's download menu in the saved report to request the review MP4. You need export permission and a Non-secure source video. Custom logos come from the report account's image Resources. Secure videos do not offer a review-overlay download.
Refresh the report before requesting the download to include its latest results. Follow preparation in Downloads; you can continue working while the video renders.
Export video dial data
Raw Excel downloads include Video Dial Results and Video Dial Raw Data sheets. SPSS bundles include a separate Excel video workbook. See Video dial data in Excel and SPSS for worksheet contents and download choices.
