Lighter Journeys: journey and task effort analysis
Jump to the moments worth a closer look in a task recording.
For UX researchers, product managers and documentation teams who watch recordings of people using their product. Upload a narrated screen recording and Lighter Journeys times the task, transcribes what was said, and marks where the narration turns negative and where the screen switched apps, with links to each moment.
Lighter Journeys is in a private preview: join the waitlist and we will invite you. Invited? Create your account with the email address we invited.
1. Record one task, thinking aloud
Two to five minutes of one person doing one task and saying what they are doing. Start just before the task and stop right after it: time on task is the length of the recording, so a detour such as checking email counts unless you cut it in the editor. MP4, MOV, M4V or WebM with a microphone works. So far we have measured accuracy on Windows recordings only.
2. Upload the recording
A two to five minute recording usually takes four to six minutes to analyze, plus any wait in the queue.
3. Jump to the moments worth a closer look
A timeline, the transcript and highlights, each linked to its moment in the video.
What Lighter Journeys measures today
| Signal | Status | What we measured | Counts toward the Effort Score |
|---|---|---|---|
| Time on task | Measured | How long the task took. | Yes |
| Parts of the task | Beta | A rough split into parts, not yet checked against real step boundaries. | No |
| Switches between apps or windows | Beta | On about 20 hours of everyday work that one person on our team recorded on Windows at 4K, it found 193 of 423 window switches, and about 2 in 3 of the switches it reported were not real (407 of 600), about one every 3 minutes. | Yes |
| Transcript | Beta | Checked on computer-generated voices, where about 1 word in 100 was wrong. A first small check on one real voice found clearly more errors, so treat transcripts of real voices as unchecked. | Yes, as the amount of narration |
| Negative narration | Beta | Whether each line sounds positive, neutral or negative. This is about the words spoken, not a measure of how anyone felt. Checked by one person on our team, who labeled recordings of their own work without seeing its answers. In ordinary work it found about 1 in 4 of the stretches they labeled negative (5 of 19), and 4 of the 9 stretches it marked negative were not. On two takes of a test site, one with built-in friction and one with it fixed, it found about 3 in 10 negative lines (11 of 35) and marked none wrongly. It reads the words, not the tone of voice. | Yes |
| Scrolls | Beta | Shown in every report with its measured accuracy. | Yes |
| Clicks | Beta | Click detection is still being improved and does not count toward the Effort Score. For accurate clicks, add Camtasia's click rings to the recording. | No |
| Typing | Beta | Typing detection is still being improved and does not count toward the Effort Score. | No |
The Effort Score is experimental too: a relative indicator for comparing sessions of the same task, not a validated measure of workload. It combines time on task, the effort of doing (scrolls, context switches and any cognitive effort labels people add), negative narration and the amount of narration. Clicks and typing do not count toward it.
Apart from the transcript check on computer voices and 6 scripted sessions in the typing figures, every accuracy figure here comes from recordings one person on our team made on Windows, on a 4K screen and a 1440 x 1080 screen. Other people, computers and screen sizes (including 1080p and 2560 x 1440) have not been checked yet.