Intent Detection
What the customer asked for and what the agent promised, returned as two separate lists because they are two different obligations and only one of them is yours to keep.
What each side meant
reading the transcript- Invoice reissued at the agreed rateprimary
- Credit rather than a carry-over
- Renewal conversation
- Reissue the invoice today
- Raise a credit note
- Book Thursday and send the terms
Product figures from the platform’s own defaults - not customer averages
How it works.
Two lists, and they are never merged
Customer intents on one side, agent commitments on the other, both as short labels rather than paragraphs. A single dominant intent is named separately and stored beside them, so a call can also be counted by the one thing it was mostly about.
A disposition is proposed from a closed set
Eleven codes: interested, not interested, callback requested, appointment set, voicemail left, wrong number, do not call, no answer, follow up needed, completed, other. A readable phrase comes with it and that phrase is not constrained at all.
The proposal stays a hint and does not click
It appears only while no disposition has been set, and setting one hides it. Your own catalogue is what a rep picks from, because those eleven codes are the model's vocabulary rather than your reporting categories.
This same block carries the writeback
Proposed column edits and proposed follow-ups arrive inside this dimension rather than beside it, which is why leaving it out of a run produces no review queue at all. A sentence of notes sits underneath the lists.
One moment in every read.
Every read passes through the same seven. Intent Detection is the lit one, and everything either side of it is a different page in this category.
- 01Source
the call or the thread it reads
- 02Transcribe
audio into words, with speakers
- 03Read
the pass over the whole of it
- 04Judge
the score, the sentiment, the intent
- 05Extract
the fields and follow-ups pulled out
- 06Write
what lands back on the record
- 07Review
a person checking the machine
The specifics.
7 facts- Returns
- A list of customer intents, a list of agent commitments, and one dominant intent
- Disposition
- One of eleven categories plus a free phrase. Neither is written onto the call
- Applying it
- By hand. Nothing maps those eleven codes onto the codes your floor uses
- Shown
- Only while the call carries no disposition. Choosing one removes the hint
- Also carries
- The proposed column edits and the proposed follow-ups. They are fields of this block
- Shape
- Free JSON on the analysis row. Its presence is checked, its interior is not
- Not the same as
- Labels and a code. The summary answers the same call in readable prose
What it reads, and what reads it.
Nothing here invents its input. These are where the material comes from, and where the verdict goes afterwards.
More in Intelligence
12 capabilitiesAI that proposes edits to the record - an insight becomes a field once you accept it.
One to six sentences on the record covering why the call happened and what was agreed, so nobody reading the list has to open the audio to find out.
Positive, neutral or negative on every analysed call, stored in a column of its own so the ones that went wrong are a filter rather than a listening exercise.
Chosen words rather than a fixed list for how the agent sounded, how the customer did, and the conversation overall - the manner behind the sentiment.
Communication, resolution, satisfaction and compliance scored nought to ten beside a separately judged overall, on every call that clears the analysis gates.
Compliance scored nought to ten against what the vertical says matters, because no phrase list exists to recite. A missing reading is excluded, never passed.
Four of six speech paths return the channel each side was recorded on, so the transcript knows who spoke rather than guessing, and one control swaps a pair.
Collections, healthcare, insurance, admissions and eleven more - each shaping what the analysis looks for without letting anyone break the output shape.
Threads summarised and read for intent and urgency, but only when a participant matches a record, so the model is never called on mail that is not yours.
Draft from an instruction or rework a message you have, landing in the composer with a suggested subject and going nowhere at all until you send it.
Describe what you track or paste a spreadsheet's columns, and get a named field group back with types, options and placeholders to approve one at a time.
A sentence becomes a grid filter and a sort order with the reading explained back, on the toolbar of every grid that carries a column menu of its own.
OpenAI, Azure, Anthropic, Gemini, Bedrock and ten more, bound per purpose with an ordered fallback chain, a circuit breaker and a cost line on every attempt.
The rest of the platform.
Five more categories, all on the same record and the same bill. Each card names three of its capabilities, so you can tell from here whether it is worth opening.
Intelligence
See intent detection on your own floor.
Thirty minutes, your numbers and your data. We will set intent detection up live and you can decide from the thing itself rather than from this page.
14-day trial · no card · migration included