Every conversation, read against criteria you control.
Pitch turns recordings and transcripts into versioned, reviewable conversation intelligence. Every finding keeps the call, rubric version and evidence that produced it attached.
The record
Recording, transcript and analysis stay one explainable object.
A recording enters one governed pipeline, receives a transcript, and is evaluated against the rubric selected for that call type. Operators can follow the result back to the words that support it instead of trusting a detached score.
Discovery call · Rubric v4
Objection handling: evidence attached at 18:42–19:16
Every analysis records the rubric version that read the call, so an answer from March remains explainable in December.
The criteria
Publish a rubric; never rewrite the past.
Publishing mints a version. Route different call types to different rubrics, test a candidate change against real calls before activation, and keep historical analyses pinned to the criteria that actually produced them.
The coaching loop
Coach from the conversation, not from a floating number.
Pitch groups the moments that need attention, exposes the transcript evidence behind each judgment, and gives managers a stable review trail without turning a model's opinion into an unchallengeable fact.
Conversation intelligence that can show its work.
Every module is part of one platform. You get all of it, and the tiers differ by scale rather than by which capabilities are switched on.