Skip to main content
All CollectionsYour meetings in tl;dvRecordings
Speaker insights for single meetings
Speaker insights for single meetings

Understanding your meeting analytics

Suellen Lorga avatar
Written by Suellen Lorga
Updated over 2 months ago

Speaker insights give you a simple, by-the-numbers look at how your calls are going so you can improve your sales calls, or guide your team towards the perfect close. πŸ˜‰

If you're on the Pro plan, you'll find speaker insights alongside your meeting recordings:

Note: If your role is in sales, this section will be expanded by default. You can always collapse or expand it with the arrow in the top right:

Who gets speaker insights?

  • Speaker insights are calculated for the members of your team in the meeting, not the external participants (leads) this is based on email domain.
    ​

  • If you use tl;dv with a public domain like gmail, you may also see insights for the other participants, but you can feel free to ignore the benchmarks for them πŸ‘

Who can see speaker insights?

  • If you use tl;dv with a private company domain, speaker insights are only shown to people from within your organization.
    ​

  • If you use tl;dv with a public domain like gmail, speaker insights are only shown to people with admin or edit access to a meeting.

What do the different insights mean?

  • Talk time - The percentage of a meeting's duration that you spent talking.
    ​E.g. 20 minutes talking in a 30 minute meeting = 66% talk time.
    ​

  • Questions/h - The number of questions asked per hour.
    ​E.g. 10 questions asked in a 30 minute meeting = 20 Questions/h.
    ​

  • Longest monologue - The longest uninterrupted period of time that you spent talking.
    ​

  • Filler words/min - The number of filler words used per minute.
    ​

  • WPM - The average number of words spoken per minute (speaking speed).

Talk time

This value helps you ensure that you (or your sales team) are listening more, and speaking less. Aim for a ratio somewhere around 40:60 with 60 percent of time occupied by the lead speaking.

Questions/h

This is the number of questions each participant asked per hour (Calculated based on the number of questions detected by tl;dv's AI and the meeting's duration). - Questions help keep your leads engaged and participating in the call, but too many questions can be detrimental so aim for between 10 and 14.

Longest monologue

This value is the longest uninterrupted time that you spent speaking, the max here should be 2 and a half minutes, to allow your leads time to interject or ask questions.

Filler words/min

Filler words are words we use like a crutch, that add little value to a meeting. There's no exact list as these are identified by AI, across 20+ languages but some English examples are:

  1. "Umm"

  2. "Ah"

  3. "Uh"

  4. "You know"

  5. "Like"

  6. "So"

  7. "Actually"

  8. "Basically"

  9. "Well"

  10. "I mean"

  11. "Okay"

  12. "Right"

  13. "You see"

  14. "Sort of"

  15. "Kind of"

  16. "Er"

  17. "You know what I mean"

  18. "And uh"

  19. "I guess"

  20. "Anyway"

The lower the better here, but nobody's perfect so you can get away with 0.6 to 3. πŸ˜‰

WPM

This is the speed at which you're speaking, averaged across the call at words per minute. - This calculation takes into account (and removes) any periods of silence.

Important: AI based Transcription is generally 95-99% accurate, so keep this in mind when reviewing the report.

Show who was speaking on the video timeline πŸ’‘

For a clear visual breakdown of who was speaking when, switch on "Show speakers on video timeline". Each participant is colour coded, so you can see how the conversation was spread out.

This may not be a concrete metric, but it's one of the best indicators of how a meeting progressed.

Ask yourself: Was it a balanced back and forth or more of a lecture?

Not seeing speaker insights?

If you don't see speaker insights for a meeting but you do see a transcript, you can trigger insights generation manually.

Did you know you can analyse speaker performance across multiple meetings? Learn how!

Did this answer your question?