Skip to main content

​
Details

A Sales Coach is a specialized role for Inbound Agents that simulates a potential customer during a sales conversation. The AI plays the prospect, and you play the sales rep. It is not a sales rep, it’s the voice on the other end of the call, portraying different buyer profiles to help you train, test, or demonstrate sales interactions. The AI behaves like a real customer by:
  • Responding to persuasion techniques
  • Displaying varied interest levels and buying styles
  • Asking realistic product-related questions
  • Raising common objections
  • Reacting to different sales tactics
After each role play, an AI evaluator reviews the transcript and scores the rep against a rubric you control. Scores and structured feedback are sent to the trainee and your distribution list automatically. This tool is ideal for:
  • Training new reps in controlled simulations
  • Stress-testing sales scripts and objection handling
  • Demonstrating call performance to stakeholders or clients
  • Offering sales coaching services to other businesses via thinkrr
All configuration fields define the simulated customer, not the company, product, or sales rep. The more realistic the persona, the more effective the training. Other members and admins on your agency account can run role plays too, so the whole team can practice with the same prospects and the same scoring. Results can be reviewed by teammate.

​
Starting a Role Play

The Sales Coach appears as a card in the Inbound Agents section of the Home screen, labeled Role play and practice selling. The card has two buttons: Start a role play and Past role plays.
1

Open the role play panel

  • From the Home screen, click Start a role play on the Sales Coach card.
2

Choose a prospect

  • Pick a prospect to practice with. Each one shows its persona and a difficulty badge, such as Moderate or Aggressive.
  • Click New prospect to create one, or click Edit next to a prospect to change it.
3

Start the session

  • Click Start session with the prospect you selected to begin the role play.

​
Creating a Prospect

The first time you create a prospect, and any time you click New prospect, a six-step setup walks you through building one.
1

Choose a scoring framework

  • Under How should we grade practice calls?, pick the framework used to score your sessions. General Sales is recommended for most teams.
  • Click Choose a different framework to see all options, or Back to recommended setup to return.
  • Choosing a framework creates a scoring package for your account. You can edit it later in the Rubric tab. See Rubric and Scoring.
2

Describe what you are selling

  • Enter a Product summary. Prospects use this context during the role play. Aim for 70–80 words, and put extra detail in a Knowledge Pack.
  • To draft it with AI, open Write with AI, describe your product in a few lines, and click Generate summary. You can edit the result.
3

Choose a prospect persona

  • Pick the persona that sets the buyer’s attitude and objections. See Personas for the full list.
4

Set the name, age, and voice

  • Enter a Prospect name.
  • Select an Age bracket and a Language.
  • Under Select voice, search or sort the list and use Listen to preview a voice.
5

Set the difficulty

  • Choose Mild, Moderate, or Aggressive, then click Create prospect. See Difficulty.
6

Start practicing

  • The Ready to practice screen summarizes the prospect, persona, difficulty, and framework.
  • Click Start practice to begin.

​
Scoring Frameworks

Every framework totals 100%.

​
Personas

The persona sets the buyer’s attitude and the objections they raise.

​
Difficulty

  • Mild – Cooperative and open. Good for building confidence and testing new scripts.
  • Moderate – Asks tough questions and needs convincing. The everyday sales call.
  • Aggressive – Resistant, skeptical, and raises objections early. For seasoned reps who want a real challenge.

​
Configuring a Prospect

Each prospect is an agent. Click Edit next to a prospect in the Start a role play panel to open its Agent Configuration screen. At the top you’ll see the prospect’s name, its Sales Coach role, and its status. The screen has six tabs: Agent Training, Actions, Web Agent, Knowledge Packs, Grades and Scores, and Rubric.

​
Credits and Model Tier

At the top of the Agent Configuration screen, you’ll see:
  • Estimated credits per minute: An estimate of how many credits the prospect consumes for each minute of a role play, based on how it is configured.
  • Available credits: Your current credit balance.
  • Using: The model tier the prospect runs on. Click Manage to change it.
Click Manage to open Select Agent Model Tier and choose Basic, which is reliable and efficient for shorter, linear conversations, or Premium, which offers deeper reasoning built for longer, more complex conversations. Role plays consume credits. For details on credit types and consumption order, see Credit Pricing.

​
Agent Training

The following input fields define how the simulated customer behaves. Identity – Basic identity of the simulated customer.
  • Customer Name – Simulated name of the customer.
  • Customer Age – Age of the simulated customer.
  • Agent Title
  • Business Name – The business your live sales rep is representing.
  • Service Type – The type of service the simulated customer is inquiring about.
Product Context – A short product anchor of 80 words max. Full detail belongs in a Knowledge Pack.
  • Product Summary
Objection Categories – The objection types the AI customer will raise. Use your own industry-specific categories, for example a Change category with “If it ain’t broke…” and a Risk category with “What if this goes wrong?” Customer Behavior – Controls the customer’s tone, resistance, and decision-making style.
  • Personality Traits
    • Friendly but cautious
    • Direct and Time Conscious
    • Indecisive but enthusiastic
  • Initial Attitude
    • Enthusiastic and eager
    • Cautiously optimistic
    • Neutral but willing
    • Slightly skeptical
    • Reserved and needs convincing
  • Interest Level
    • Moderately interested, information gathering
    • Slightly interested, just browsing
    • Highly interested, price sensitive
  • Solution Receptiveness
    • Highly receptive
    • Open minded but needs validation
    • Cautious but willing
    • Skeptical but convincible
    • Resistant but listening
  • Price Sensitivity
    • Very price sensitive
    • Moderately price conscious
    • Value-focused
    • Budget driven
    • Premium-oriented
  • Decision Style
    • Quick and decisive
    • Methodical and researching
    • Emotional and intuitive
    • Collaborative and consulting
  • Communication Style
    • Direct and to-the-point
    • Detailed and thorough
    • Casual and conversational
    • Formal and professional
  • Difficulty Level
    • Easy
    • Medium
    • Hard
Persona Behavioral Parameters – Controls how the customer resists, escalates, and reacts under pressure.
  • Resistance Style
  • Reaction to Pressure
  • Vocabulary Level
  • Escalation Pattern
Interaction Context – Scenario context, interaction type, and call objective.
  • Personal Background – Short description of the customer’s situation or lifestyle.
  • Lead Source
    • Social Media advertisement
    • Google search
    • Friend or family recommendation
    • Drive by or walk by
    • Corporate wellness program
    • Local community event
    • Current member referral
  • Interaction Type
    • Outbound call
    • Inbound call
    • Sales pitch
    • Objection handling
  • Interaction Setting
    • Introductory call
    • Lead nurture call
    • Sales closure call
    • Service upgrade call
  • Call Objective
We recommend not overthinking the instructions. Train your agent the way you would brief a roleplay partner, not with a rigid script, but with traits, context, and likely reactions to simulate. Remember, this is Generative AI. The more flexible and realistic the persona you provide, the better the simulation.
A Knowledge Pack allows your agent to access and reference structured company information during conversations. It can be used across both inbound and outbound projects.Depending on your plan, you can create one or more Knowledge Packs and attach them to specific agents as needed. Each Knowledge Pack uses one slot.Supported plans:
  • Agency Lite: 1 Knowledge Pack slot
  • Agency Unlimited: 2 Knowledge Pack slots
Additional slots can be unlocked from the Knowledge Packs page in the left-hand sidebar.

​
Steps to Add a Knowledge Pack:

  1. In your inbound agent configuration screen, scroll down and click “Knowledge Packs.”
  2. Click “Create New Knowledge Pack” to launch the builder.
  3. Choose one or both of the following sources:
    • Upload files in .md, .csv, .tsv, .xlsx, .xls, .pdf, .txt, or .docx format
      • Up to 25 files per Knowledge Pack, 50MB per file
    • Enter URLs to let thinkrr scan web pages for reference content
  4. Once created, you can edit, delete, or refresh your pack from the right-hand menu.
    • To disable access temporarily, simply turn off the pack instead of deleting it.
    • You can re-enable the same pack at any time without losing the content.
🚫 Note: Sub-accounts cannot access Knowledge Packs created by an agency’s main account.

​
Actions

The Sales Coach role can send notifications about training sessions but does not handle real bookings or CRM actions.
  • Notify The Rep
    • Enter up to 10 email addresses.
    • Each notification includes:
      • Simulated caller profile
      • Conversation summary
      • Transcript
      • Call recording (if enabled)
  • Extract Custom Fields
    • Captures structured data from each call and stores it against the contact’s record in thinkrr.
    • Toggle Extract custom fields on, then click Action setup to select the fields to extract.
    • For full setup instructions, see Abilities & Actions.

​
Web Agent

Use the Web Agent tab to configure how the prospect appears on the web. See Web Widget.

​
Prospect Settings

Click the Settings button at the top right of the agent card to adjust the prospect’s settings. They are organized into three tabs: Agent Settings, Call Settings, and Voice Settings.
  • Agent Settings
    • Agent Name – Name the simulated customer persona (e.g., “Skeptical Prospect” or “Premium Buyer”).
    • Language & Ethnicity – Choose the language and voice traits to match the persona.
    • Agent Voice – Use the Listen button to preview and select a voice.
    • Phone Number – Assign a number if phone simulation is required (Inbound Agents must have unique numbers).
  • Call Settings
    • Include History – Option to let the Sales Coach “remember” past simulated calls.
    • Noise Isolation Mode – Filters out background voices.
    • Maximum Call Duration – Caps call length (set in minutes).
    • End Call After Silence – Ends the call automatically after a set silence period.
    • Background Sound – Add ambient noise (e.g., busy office).
    • Background Volume Slider – Appears if background sound is enabled.
  • Voice Settings
    • Loudness Level (volume) – Adjust how loud or soft the simulated customer speaks.
    • Responsiveness – Decide how quickly the persona replies.
      • Less responsive: Pauses more before answering.
      • More responsive: Reacts instantly for fast-paced simulation.
    • Interruption Sensitivity – Controls how easily you can cut off the agent mid-sentence.
      • High sensitivity: Interruptions cut in immediately.
      • Low sensitivity: Persona will usually finish speaking first.

​
Grades and Scores

Every role play is scored automatically. You can review results in two places:
  • Past role plays on the Sales Coach card on the Home screen
  • The Grades and Scores tab in a prospect’s Agent Configuration
Both show the same information:
  • Session results – The number of scored sessions, with a rows-per-page selector.
  • Teammate / contact filter – In Past role plays, filter by All teammates or a specific person to see who ran each role play.
  • Each session – The person’s name, email, and the date and time, plus the overall score as a percentage and a Call details button.
Expand a session to see two sections:
  • Rubric breakdown – Each dimension with its score out of 10, its weight, an explanation of the score, and the criteria it was judged against.
  • Feedback summary – Coaching feedback in the sections defined in your Feedback Format, for example opening strength, discovery quality, and the closing attempt.
Click New role play in Past role plays to start another session.

​
Rubric and Scoring

After each practice call, an AI evaluator reviews the transcript and scores the rep against the rubric you define in the Rubric tab. Scores and structured feedback are sent to the trainee and your distribution list automatically. The Rubric tab lists your scoring packages. Each package shows its number of dimensions and feedback sections, its total weight, and whether it is Selected or Not selected. The selected package is the one used to score role plays. Use the three-dot menu on a package for more options, and click the arrow to open it. Click Add package to create a package. Template packages, such as General sales and SaaS / tech, are available as a starting point.

​
Rubric

Open a package and select the Rubric tab to define how sessions are scored.
  • Weight coverage shows your total weight. Weights use values between 0 and 100, and the total must be exactly 100 to save.
  • Name – The package name.
  • Dimensions – Click Add dimension to add more. Each dimension has:
    • Dimension name
    • Weight (%)
    • Criteria – What the evaluator looks for.
For example, the General sales template scores Communication clarity (25), Objection handling (30), Rapport building (20), and Closing technique (25).

​
Feedback Format

Select the Feedback Format tab to control how coaching feedback is written. The post-call evaluator follows this structure when generating feedback from the transcript and rubric scores.
  • Start from a template:
    • Standard coaching – Strengths, areas for improvement, and next steps
    • Brief summary – A score plus one action-focused paragraph
    • Detailed per-dimension – A score and narrative for each dimension
  • Instructions – The feedback sections. Click Add section to add more. Each section has:
    • Section name
    • Guideline – How the evaluator should write it. For example, “What you did well” highlights two to three specific moments from the transcript.
  • Tone Requirement – For example, constructive and encouraging.
  • Compliance Notes – For example, flag only.
Click Save to apply your changes, or Discard to cancel.

​
Notes & Considerations

  • This role is for Inbound Agents only.
  • The Agent simulates a customer, not a company rep.
  • Use role plays to train and certify human reps or validate outbound scripts.
  • You can set up multiple prospects to simulate different customer types, personas, and difficulty levels.

​
FAQs & Troubleshooting

​
General Questions

The Sales Coach is only available as an Inbound Agent, but it simulates outbound-style customer calls to train your team.
It acts like a potential customer, responding based on your inputs, asking questions, showing interest or hesitation, and behaving like a real prospect.
Yes. Choose the Aggressive difficulty and a persona like The Wounded Buyer or The Value Squeezer, then fine-tune fields like Initial Attitude, Difficulty Level, and Objection Categories in the prospect’s Agent Training tab.
Other members and admins on your agency account can run role plays. In Past role plays, you can filter results by teammate to see who ran each session.
Yes. Role plays consume credits based on how the prospect is configured. The estimated credits per minute is shown at the top of the prospect’s Agent Configuration screen.

​
Configuration

An AI evaluator reviews the transcript after each role play and scores the rep against the selected scoring package in the Rubric tab. Scores and feedback are sent to the trainee and your distribution list automatically.
Yes. Open the Rubric tab to edit a package, or click Add package to create your own. Set the dimensions, criteria, and weights, then mark the package as selected to use it for scoring.
Dimension weights must total exactly 100 to save. The weight coverage bar at the top of the package shows your current total.
Yes. Open a package, select the Feedback Format tab, and edit the sections, guidelines, tone requirement, and compliance notes. You can also start from a template.
Click Start a role play on the Sales Coach card, then click Edit next to the prospect to open its Agent Configuration screen.

​
Usage and Results

Click Past role plays on the Sales Coach card, or open the Grades and Scores tab in a prospect’s Agent Configuration screen. Expand a session to see the rubric breakdown and feedback summary.
Choose a harder difficulty and make sure the prospect includes a realistic Difficulty Level and specific Objection Categories.
Try refining Price Sensitivity and providing more context in Personal Background. Adding a price-related persona, like The Value Squeezer, also helps.
Add more depth to Communication Style, Decision Style, and Personal Background, and include nuanced objections or goals.
For additional questions or guidance, try using our Virtual Support Agent! Available 24/7 to help resolve most issues quickly at thinkrr.ai/support. If you still need assistance, visit our support site at help.thinkrr.ai and submit a Ticket or contact our team directly at hello@thinkrr.ai.