Every Intelligence Field, Explained
What each result means, and how far to trust it.
A processed meeting produces the results below. They are all worked out from the transcript in a single pass, so they are consistent with each other — the sentiment reading and the summary are describing the same reading of the conversation.
A field with nothing in it is omitted rather than shown empty, so a shorter screen means less was found, not that something failed.
The three summaries
Meeting Summary
Two to three sentences
optional
The factual record: what was discussed, what was decided, what the outcome was. Written in the third person and deliberately neutral, because this is the text that goes onto the contact's record as a note where it will be read months later by someone who was not there.
Team Debrief
Two to three sentences
optional
The same meeting written for you — outcomes, key decisions and where the relationship now stands. Warmer than the summary and more use at a glance. This is what your debrief email carries.
Summary for Attendee
A short message
optional
Written as you, to the person you met: using their name, referring to specific things they said, and moving naturally to what was agreed. It reads like a follow-up from a person rather than a record of a meeting, because if you switch the attendee summary on, this is the message they receive.
Tip: Read the Summary for Attendee before you switch attendee summaries on. It is the only one of the three that leaves your business, and reading a couple of real examples tells you more about whether you want it sending automatically than any description can.
The scores
Sentiment
positive, neutral, negative or mixed
optional
The overall tone Zyntro read in the conversation, shown as a badge with a percentage beside it. Mixed is a real and common answer rather than a failure to decide — plenty of good meetings contain a frank objection.
Engagement
A percentage
optional
How involved the participants were, shown as a filled bar. Roughly: 70% and above reads as high, 40% to 70% as moderate, below 40% as low.
Participation
Balanced or Uneven
optional
Whether the talking was shared out or dominated by one person, shown as a badge under the engagement bar. Uneven on a discovery call is the more useful reading of the two — it usually means you were doing the talking.
Most active
A speaker name
optional
Who spoke most, named beside the participation badge. Together with the badge this is what makes the engagement percentage actionable: a number on its own tells you little, whereas knowing one person carried the conversation tells you what to change.
The reasoning
A short explanation of each score
optional
Underneath the two scores, Zyntro explains how it read the tone and the participation. Worth reading before taking either percentage at face value — it often names the specific exchange that moved the reading.
Tip: Read the participation badge before the engagement number. A 45% engagement score means something quite different on a meeting where the talking was shared than on one where a single person spoke throughout. The badge and the most-active speaker name are what turn the score into something you can act on.
What was said
Key Highlights
A list
optional
The points worth remembering, each attributed to a speaker where Zyntro could tell who said it. This is the fastest way back into a meeting you half-remember.
Action Items
A list
optional
Specific things to be done, each with an assignee and a deadline where those were actually mentioned in the conversation. Nothing is invented: an action with no owner is one nobody claimed out loud, which is itself useful to see.
The follow-up recommendation
Topic
Text
optional
What the next conversation should be about.
Timing
Text, such as "Within 3 days"
optional
How soon, based on what was said in the meeting rather than a fixed rule.
Reason
Text
optional
Why a follow-up is recommended. Worth reading — it often names the specific thing left hanging.
Info: The recommendation appears whether or not you switched the follow-up setting on. Working out the next step is part of reading the meeting, so it is always shown here.
What the setting adds is a reminder: with it on, Zyntro turns the recommendation into a dated notification addressed to you, filed against the contact and linking back to this meeting. One per meeting, so re-processing does not stack them up.
Relationship insights
Rapport Level
strong, moderate, weak or unknown
optional
How the relationship read in this conversation. Unknown is an honest answer on a short or purely transactional call.
Concerns Raised
A list
optional
Objections, hesitations and worries the person expressed. The most actionable field on the screen — these are the things to have an answer for next time.
Interests
A list
optional
What they responded well to, shown as tags. Useful for deciding what to lead with.
Next Steps
A list
optional
What was explicitly discussed as happening next. Distinct from the action items: these are things said in the conversation, not tasks assigned.
Results that go to the contact, not the meeting
| Result | What it is | Where it goes |
|---|---|---|
| Pain points | Problems the person described, each with a label, a description, the evidence from the conversation, and a strength of high, medium or low. | The contact's profile |
| Desires | What they want to achieve, in the same shape. | The contact's profile |
| Motivations | The underlying drivers behind what they want. | The contact's profile |
| Fears | What they are worried about going wrong. | The contact's profile |
| Inferred details | Specific facts Zyntro picked up — a job title, a timezone, the kind of organisation they are. Applied only where it was confident. | The contact and entity records |
Important: Two of the most useful results are not on the meeting screen at all. What Zyntro learned about the person — their pain points, what they want, what is driving them — and any details it inferred about them are written onto the contact's record, not shown against the meeting.
So a meeting screen with no sign of them does not mean they were not produced. Open the contact to see them.
Warning: Each result is a reading of a conversation, not a measurement of one. The summaries and highlights are reliable because they are drawn from what was said. The sentiment percentage, the engagement percentage and the rapport level are interpretations — useful for spotting a pattern across many meetings, and not worth arguing with on any single one.
The evidence shown with each pain point and desire exists for this reason: you can see the words it was drawn from and judge for yourself.
