Company reference
How GhostRep uses company signals.
GhostRep gets sharper the longer your company uses it. The platform learns from your conversations, role plays, coaching notes, job context, and the priorities a sales manager sets — and turns that into coaching, practice, summaries, and training tied to your business.
6 signal groups
Inputs the system learns from
5 surfaces
Outputs shaped by your company
After manager review
AI outputs reach reps
No
Raw transcripts exposed by default
The premise
A generic sales tool cannot coach your team. Home improvement sales does not look like SaaS sales, and a roofing rep in a hail market does not have the same conversations as an HVAC rep selling a slow quote.
GhostRep is designed to learn the texture of your company: the products you sell, the objections you actually hear, the way your manager wants pricing handled, and the moments where deals get won or lost on your team. That shows up as sharper coaching and outputs that read like they came from someone who knows your business — because they did.
Company signals
What the system actually learns from.
Nothing exotic. The day-to-day signal a contractor sales floor produces while it operates.
01
Echo sessions and field conversations.
When reps run Echo on an in-home appointment or canvassing route, the platform captures the conversation and turns it into structured signal: which objections came up, where the deal stalled, how the rep handled price pressure, what the homeowner asked, and which moments mattered most.
02
Role play sessions.
Reps practice against AI homeowner scenarios across roofing, HVAC, solar, windows, doors, and other home improvement contexts. The system records what the rep tried, where they got stuck, and which patterns repeat across the team.
03
Canvassing and appointment capture.
Activity at the door, in the appointment funnel, and in follow-up shows where reps are spending time and where conversations actually turn into appointments and sat demos.
04
Coaching notes and manager priorities.
Managers tag what they want reps to work on. Those notes — combined with the priorities a sales leader sets inside AI Sales Coach — direct where coaching focuses next.
05
Job Intel and CRM context.
Job notes, customer questions, inspection findings, and the context a manager attaches to a job become part of the picture the rep sees before the appointment and what the system reasons about afterward.
06
AI coach settings and company playbooks.
Owners and managers set the coaching style, the priority objections, the products and pricing structure, and the close standards. Those settings shape what AI Sales Coach surfaces and how Training Studio builds material.
What it produces
Outputs shaped by your company.
Each is a draft or recommendation a manager can review and approve. None of it is automatic policy.
Coaching
AI Sales Coach summarizes what is happening across the team, flags trends, and recommends what each rep should work on next — based on actual sessions and notes, not assumptions.
Practice
Role Play scenarios get sharper as the system sees real objections, real customer questions, and real pressure points the team is running into in the field.
Summaries
Echo and Job Intel turn raw conversation and job context into concise summaries managers can scan — so the manager spends time coaching, not transcribing.
Training material
Training Studio drafts company-specific training modules from the playbooks, objections, and priorities a company actually uses. Managers approve before anything ships to reps.
Manager-ready outputs
Coaching recommendations, recap notes, and rollup reports come pre-structured so managers can review, edit, approve, and act — without rebuilding from scratch.
Manager control
Owners and managers decide what gets used.
Coaching outputs, training modules, and recommendations are starting points. A human signs off before anything becomes the standard for the team.
- →
Managers and admins control what data is reviewed, who can see it, what gets approved for team-wide use, and what gets exported.
- →
Training Studio outputs, AI coaching recommendations, and summaries are drafts, not automatic policy. A human approves before anything is rolled out to reps.
- →
Raw transcripts are not broadly exposed by default. Access to underlying session content follows your organization's permission structure.
- →
AI outputs can be wrong. GhostRep treats them as starting points for a human reviewer, not as final answers.