# Create a Personalized Bot
Source: https://docs.getzell.com/guides/bots
Build custom AI bots with unique personas for training and practice scenarios
## Overview
Create personalized AI bots that simulate real customer interactions for training your team. Each bot can have a unique persona, communication style, and scenario focus to provide diverse practice experiences.
## Creating a Persona
A persona defines the character and background of your bot. This is the first step in creating a realistic training experience.
### Persona Fields
When creating a persona, you'll define:
* **Job Title:** The role of the person (e.g., "VP of Sales", "IT Manager")
* **Industry/Company Description:** The business context (e.g., "B2B SaaS in healthcare")
* **Company Size:** Startup, Small, Medium, Large, or Enterprise
* **Age Range:** Optional (e.g., "35-45")
* **Location:** Optional geographic location
* **Objections:** Common concerns or pushbacks they typically raise
* **Problems to Solve:** What challenges they're facing
* **Current Solutions:** What they're currently using
* **Goals:** What they're trying to achieve
* Optionally assign the persona to a group for organized access
## Creating a Bot
Once you have a persona defined, you can create a bot that uses it for training scenarios.
### Bot Configuration Steps
Go to **Team Management** → **Bots** in the main navigation.
Click **Create Bot** to start configuring your new training bot.
Choose from your existing personas. Group labels appear beside each persona so you can quickly identify the right one. The persona provides the background and context for how the bot will behave.
Create personas first, then build bots that use them.
Set up the specific details for this bot:
* **Name:** A descriptive name for the bot
* **Scenario/Context:** The specific situation for this training
* **Voice Settings:** Configure how the bot sounds and speaks
* **Evaluation:** Link to a scorecard for assessment
* Click **Save** to create your bot
* Test the bot to ensure it works as intended
* Make adjustments as needed
Make the bot available to specific groups for training access.
## Bot Use Cases
Create bots that present common objections to help reps practice responses.
Simulate prospects with complex needs requiring thorough qualification.
Build bots that ask detailed technical questions for solution engineers.
Practice price negotiations and deal structuring with tough buyers.
Create frustrated customers for support team training.
Build friendly bots for new hire practice in a low-pressure environment.
## Best Practices
Base personas on actual customers or prospects you've encountered in your market.
Include concrete details in persona fields—specific objections, problems, and goals.
Connect bots to scorecards so team members know what they're being evaluated on.
Always test new bots yourself before assigning them to your team.
Use groups to control which personas and bots are available to different teams.
Regularly update personas based on market changes and new objections.
# Calls
Source: https://docs.getzell.com/guides/calls
Review your calls, coaching moments, and skill gaps from one workspace
## Overview
The **Calls** workspace brings your reviewed calls and coaching insights into one place. Use it to understand where to focus, review coachable moments, and open individual call analyses when you need more detail.
The workspace has two views:
Review focus areas, coachable moments, and skill gaps from your calls.
Browse your call history, filter calls, and open a full call analysis.
## Open Calls
Open **Calls** from the main dashboard navigation.
Use the view switcher at the top of the page to move between **Coaching Insights** and **Table View**.
## Coaching Insights
Coaching Insights is designed for focused self-review. It summarizes coachable moments from your calls and groups them by skill area.
At the top of the view, you can see:
* **Moments:** Total coachable moments found in the selected calls
* **Reviewed:** How many moments you have already reviewed
* **Focus Areas:** The number of skill areas with coaching opportunities
* **Review Progress:** Your reviewed moments compared with all available moments
### Filter Insights
Use the filters to narrow what you are reviewing:
* **Scorecard:** Focus on calls evaluated with specific scorecards
* **Call Source:** Show all calls, real calls, or simulations
* **Date Range:** Limit insights to a specific time period
### Review Focus Areas
Coaching Insights shows your focus areas in a single list. Select a focus area to see the related snippets from your calls beside it.
Use this view when you want to work through one skill at a time. Each focus area shows how many moments are available, and the selected focus area shows its review progress together with the matching snippets.
### Review Moments
When you open a focus area, Zell shows the relevant snippets from your calls. From each snippet, you can:
* Play the call moment
* Review the coaching feedback
* Add the call to a collection
* Vote on whether the feedback was helpful
Reviewing moments updates your progress, making it easier to see what still needs attention.
## Table View
Use **Table View** when you want to browse individual calls or open a full call analysis.
You can filter the call list by:
* **Scorecard:** Show calls evaluated with selected scorecards
* **Coaching Score:** Show all, high, medium, or low coaching scores
* **Date Range:** Limit the list to a specific time period
The table includes the call date, customer or agent, duration, scorecard category scores, **Coaching Need**, and unread comments. **Duration** appears as its own column so you can scan call length without opening the full analysis.
Click a call in the table to open its full call analysis. Use pagination controls to move through larger call lists.
## Call Analysis
The full call analysis page shows the call recording, transcript, summary, coaching feedback, and evaluation details for an individual call.
### Review Video Recordings
For calls with video, hover over or focus the player to reveal its playback controls. Use the backward and forward buttons to jump 10 seconds in either direction.
To resize the player, drag its lower-left resize handle. You can also focus the resize handle and use the left and right arrow keys to make the player smaller or larger.
For real calls, the header also shows customer and call context when available, including the contact name or email, company, call date, duration, language, call type, and outcome. This makes it easier to confirm you are reviewing the right conversation before working through the analysis.
Users with the right permissions can keep real-call details up to date directly from the header:
* **Managers and team leads:** Update the call language, call type, and outcome
* **Admins:** Edit the call title and reassign the call owner
Changes are saved from the header controls. If an update cannot be saved, Zell shows an error so you can try again.
## Best Practices
Use Coaching Insights first to identify the skills with the most coachable moments.
Narrow insights to the scorecard that matches the call type or skill you are improving.
Work through a few moments at a time so coaching feedback stays actionable.
Use Table View when you need the full conversation context behind a moment or score.
# Set Up a Course
Source: https://docs.getzell.com/guides/courses
Create comprehensive training courses with modules, steps, and automated enrollment
## Overview
Courses provide structured learning paths for your team with multiple modules, text content, and agent practice sessions. Each course has an **intensity preset** that controls the number of practice calls required and the minimum score a call must reach to count.
## Intensity Presets
When creating a course, you select one of three intensity presets that define the training rigor:
| Preset | Calls per Exercise | Min. Score per Call | Best For |
| ------------- | ------------------ | ------------------- | -------------------------------------------- |
| **Light** | 2 calls | ≥6/10 | Refreshers, product updates, quick check-ins |
| **Standard** | 5 calls | ≥7/10 | Ongoing training, skill building |
| **Intensive** | 10 calls | ≥8/10 | New hire onboarding, certification programs |
Intensity presets are locked—you select a preset and cannot customize individual values. This ensures consistent training standards across your organization.
## Creating a Course
Course creation is a 2-step process: first configure the course details, then build the module structure.
### Step 1: Course Details
Go to **Team Management** → **Courses** in the main navigation and click **Create Course**.
Configure the basic course information:
* **Title:** Enter a clear, descriptive course name (minimum 3 characters)
* **Description:** Summarize what learners will achieve (minimum 3 characters)
**Example:**
* Title: "Sales Fundamentals Bootcamp"
* Description: "Master consultative selling through interactive lessons and practice sessions with AI prospects."
Choose the training intensity that matches your course goals:
* **Light:** Quick refreshers with 2 calls per exercise, each call must score ≥6/10
* **Standard:** Regular training with 5 calls per exercise, each call must score ≥7/10
* **Intensive:** Rigorous programs with 10 calls per exercise, each call must score ≥8/10
Start new hires with **Intensive** courses for thorough onboarding, and use **Light** for periodic product updates.
Choose which groups this course is designed for. This filters the available agents and users to only show those relevant to the selected groups.
**Why use target groups?**
* Focus on specific teams or departments
* Filter agents trained for those teams
* Pre-filter users when enrolling
Select which team members should be enrolled in this course. If you selected target groups, only users from those groups will be shown.
You can add or remove users later directly from the course card—no need to edit the course.
Click **Next** to proceed to building your course modules.
### Step 2: Build Modules
Click **Add Module** for each major topic in your course. Each module is a collapsible section containing related steps.
For each module, provide:
* **Module Title:** The name of this section (e.g., "Discovery Fundamentals")
* **Module Description:** What participants will learn in this module
Modules appear in the order you create them. Click on a module to expand or collapse its steps.
Inside each module, click **Add Step** and choose the step type:
**Text Step:** Educational content, instructions, or reference materials
* **Title:** Name of the step
* **Description:** Optional additional context
* **Content:** The actual text content for learners to read
**Agent Call Step:** Practice sessions with AI agents
* **Title:** Name of the practice session
* **Agent:** Select which bot learners will practice with (filtered by target groups if selected)
The number of calls required is automatically set by your course's intensity preset. For example, a **Standard** course requires 5 calls per agent step.
Alternate between text steps (learning) and agent call steps (practice) for better retention.
Click **Create** to save the course and make it available to enrolled users.
## Step Types
### Text Step
Text steps deliver written educational content to learners.
**Fields:**
* **Title:** Name of the step (required, minimum 3 characters)
* **Description:** Optional additional context
* **Content:** The actual text content learners will read
**Use Cases:**
* Introduction to concepts
* Theory and frameworks
* Best practices guides
* Scripts and talk tracks
* Reference materials
### Agent Call Step
Agent call steps require learners to complete practice conversations with AI bots. The number of calls and the minimum score per call are determined by the course's intensity preset.
**Fields:**
* **Title:** Name of the practice session (required, minimum 3 characters)
* **Agent:** Select which bot learners will practice with (required)
**Pass Criteria:**
* Each call must score at or above the preset's minimum (≥6/10, ≥7/10, or ≥8/10) to count toward completion
* Accumulate the required number of passing calls (2, 5, or 10 depending on preset)
**Use Cases:**
* Applying learned techniques
* Building muscle memory
* Handling specific scenarios
* Skill practice and assessment
## Managing Enrollment
### Adding Users to a Course
You can add users directly from the course card without editing the course:
On any course card, click **Enrolled People** to expand the enrolled users section.
Click the **+ Add User** button at the top of the enrollment list.
Search for and select the user you want to enroll. They'll be added immediately.
### Removing Users from a Course
On the course card, expand the enrolled users section.
Hover over the user row and click the **X** button that appears to remove them from the course.
Removing a user from a course does not delete their progress—if re-enrolled, their progress will be restored.
## Example Course Structure
Here's a complete example of a well-structured course:
**Intensity:** Standard (5 passing calls per exercise, each call must score ≥7/10)
**Module 1: Discovery Fundamentals**
1. Text Step: "Introduction to Consultative Selling"
2. Text Step: "The SPIN Questioning Framework"
3. Agent Call Step: Practice discovery calls (5 passing calls with "Friendly Prospect")
**Module 2: Qualifying Prospects**
1. Text Step: "BANT Qualification Framework"
2. Text Step: "Reading Buying Signals"
3. Agent Call Step: Qualification practice (5 passing calls with "Mixed Signals Prospect")
**Module 3: Presenting Solutions**
1. Text Step: "Features vs Benefits"
2. Text Step: "Tailoring Your Pitch"
3. Agent Call Step: Pitch practice (5 passing calls with "Technical Buyer")
**Module 4: Handling Objections**
1. Text Step: "Common Objections and Responses"
2. Text Step: "The Feel-Felt-Found Method"
3. Agent Call Step: Objection handling drills (5 passing calls with "Skeptical Prospect")
**Module 5: Closing Techniques**
1. Text Step: "Recognizing Buying Signals"
2. Text Step: "Trial Closes and Assumptive Closes"
3. Agent Call Step: Full cycle role plays (5 passing calls with "Ready-to-Buy Prospect")
## Managing Courses
### Viewing Courses
Navigate to **Team Management** → **Courses** to see all courses with:
* Course title, description, and intensity preset
* Number of enrolled users
* Total steps across all modules
* Overall progress percentage
* Enrolled user details and individual progress
### Editing a Course
To update an existing course:
1. Click the **Edit** icon on the course
2. Update course details:
* Title
* Description
* Intensity preset
3. Save your changes
Changes to intensity preset affect the call requirements and minimum score per call for all steps going forward. User enrollment is managed directly on course cards.
### Tracking Progress
For each course, you can monitor:
* **Enrollment count:** How many users are enrolled
* **Overall progress:** Average completion percentage across all users
* **Individual progress:** Each user's completion status and average score
* **Completed steps:** Number of steps each user has finished
* **Module completion:** Which modules users have completed
## Best Practices
Use Intensive for critical skills (onboarding, certification), Standard for ongoing training, Light for refreshers.
Begin each module with text steps explaining concepts before practice.
Alternate between text (learning) and agent calls (practice) for better retention.
Start with easier scenarios and gradually increase complexity.
Filter agents and users by group to ensure relevant practice scenarios.
Keep modules focused on one skill to avoid overwhelming learners.
# Custom Call Types & Outcomes
Source: https://docs.getzell.com/guides/custom-call-types-outcomes
Configure the call categories and result labels Zell uses for scoring, reports, and integrations
## Overview
Call types and call outcomes control how Zell classifies real customer conversations.
What kind of conversation took place, such as a cold call, demo, discovery call, renewal call, or support call.
What happened as a result of the conversation, such as a demo booked, no interest, information requested, or another team-specific result.
Zell includes default values for both concepts. Admins can add custom values when the defaults do not match the team's process.
***
## How Zell Uses Them
### Call Types
Call types are used to route calls into the right coaching and evaluation setup:
* **Scorecards** can be assigned to specific call types.
* **Default scorecards** use call type to decide which scorecard should automatically score an imported real call.
* **Bots and playbooks** can be aligned with the kind of conversation a user needs to practice.
* **Reports** can be filtered by call type to compare performance across conversation types.
### Call Outcomes
Call outcomes are used to summarize the result of a real call:
* **Real call reports** can be filtered by outcome.
* **Imported calls** from dialers and calling tools can be categorized automatically.
* **Integration mappings** can map external tags or dispositions into Zell's outcome labels.
* **Coaching reviews** can separate successful calls from calls that need follow-up.
***
## Default vs. Custom Values
Default values are available to every workspace and are read-only. Zell localizes the default labels in the app where supported.
Custom values belong to your tenant. They appear alongside the defaults in call type and outcome selectors.
Custom values do not replace defaults. If you no longer need a custom value, remove it from the custom list.
### Default Call Types
Zell's default call types include:
* Cold Call
* Demo
* Gatekeeper
* Discovery
* Check-in
* Warm Call
* Customer Care
* Proposal
### Default Call Outcomes
Zell's default call outcomes include:
* Contact Received
* Demo Booked
* Information Requested
* No Interest
* Pilot Booked
* Stakeholder Intro
* Other
***
## Add Custom Call Types
Open **Settings** → **Tenant Settings**.
Select **Call Types** from the tenant settings menu.
The default call types are shown as locked labels. They are available automatically and cannot be edited.
Click **Add call type**, enter a label, and add a description that explains when this call type should apply.
Use the same language your team already uses, such as "Renewal Call", "Onboarding Call", or "Technical Qualification".
Click **Save Changes**. The custom call type becomes available in scorecards, bots, playbooks, integrations, and report filters.
***
## Add Custom Outcomes
Open **Settings** → **Tenant Settings**.
Select **Call Outcomes** from the tenant settings menu.
The default outcomes are shown as locked labels. They are available automatically and cannot be edited.
Click **Add outcome** and enter the result label your team needs.
Write outcomes as final states, such as "Trial Started", "Budget Confirmed", or "Competitor Mentioned".
Click **Save Changes**. The custom outcome becomes available for real call categorization, reporting, and integration mappings.
***
## Map Integrations
For connected calling tools, configure mappings after you add custom values.
### Aircall
Aircall tags can be mapped to a call type, an outcome, or both. When a mapped tag is present on an imported call, Zell uses the mapped value instead of relying on automatic classification.
Go to **Tenant Settings** → **Integrations** → **Aircall**.
Click **Refresh Tags** if the tag list is outdated.
For each Aircall tag, choose the matching call type and/or outcome.
Save the mapping so future Aircall imports use your configured labels.
### Amplemarket
Amplemarket dispositions can be mapped to Zell call outcomes.
Go to **Tenant Settings** → **Integrations** → **Amplemarket**.
Refresh dispositions to pull the latest values from Amplemarket.
Match each disposition to the Zell outcome it represents.
Save the mapping so future Amplemarket imports use your configured outcomes.
***
## Limits and Behavior
* You can add up to **50** custom call types.
* You can add up to **50** custom call outcomes.
* Labels cannot be empty.
* Duplicate custom labels are ignored.
* Custom labels that exactly match default labels are ignored.
* Saving replaces the current custom list with the values shown on the settings page.
Removing a custom call type can affect scorecards, bots, playbooks, and mappings that were using it. Review related configuration before deleting values that are already in use.
***
## Best Practices
Use labels that are easy to scan in filters, tables, and mapping menus.
Do not create custom values that mean nearly the same thing as an existing default or another custom value.
Match your CRM, dialer, and sales process vocabulary so mappings are easier to maintain.
After adding a new call type, assign relevant scorecards and default scorecards so real calls are evaluated automatically.
# Engagement Score
Source: https://docs.getzell.com/guides/engagement-score
Understand how Zell measures whether coaching feedback is being reviewed
## Overview
The engagement score shows how consistently people review the coaching moments Zell surfaces from their calls.
It is shown from `0.0` to `10.0`:
* `10.0` means all counted coaching moments were reviewed.
* `5.0` means about half were reviewed.
* `0.0` means no counted coaching moments were reviewed.
Engagement is about reviewing coaching feedback. It is not a measure of call quality, call volume, or how often someone logs in.
## What Counts
A coaching moment counts toward engagement when Zell identifies it as a useful improvement opportunity. A moment is considered reviewed after the user opens or reads that feedback.
Zell focuses this score on meaningful coaching moments, so it does not count every comment or every call. Feedback that has been marked as unhelpful is also excluded.
## How To Read It
| Score | Meaning |
| ------------ | ------------------------------------------------------------- |
| `7.0 - 10.0` | Most coaching feedback is being reviewed. |
| `4.0 - 6.9` | Some feedback is being reviewed, but follow-up may be needed. |
| `0.0 - 3.9` | Coaching feedback is often not being reviewed. |
Managers can use the score to see whether the team is closing the loop on coaching, not just generating feedback.
## Why It May Change
The score can change when:
* A user reviews new coaching feedback.
* New coachable moments are found from recent calls.
* Filters change, such as date range, scorecard, call source, user, or group.
* Previously counted feedback is marked as unhelpful and excluded.
## Example
If Zell surfaces 10 counted coaching moments for a user and they review 7 of them, their engagement score is `7.0`.
If a team has 40 counted coaching moments and 30 have been reviewed, the team engagement score is `7.5`.
## Best Practices
* Review coaching moments soon after calls while the conversation is still fresh.
* Use date filters to compare recent engagement against previous periods.
* Look at low engagement together with performance scores. A low score often means the person is not seeing the feedback that could help them improve.
# Connect Gong
Source: https://docs.getzell.com/guides/gong-integration
Import your Gong calls and scorecards into Zell for AI-powered coaching
## Video Walkthrough
## Overview
The Gong integration lets you import real sales calls directly into Zell — including transcripts, recordings, and scorecards. Once connected, you can use Zell's AI coaching features on your actual conversations.
## Setup Steps
From anywhere in the app, click your avatar and select **Tenant Settings**.
Navigate to **Integrations** and locate **Gong**.
Click **Connect to Gong**. You'll be redirected to Gong's authorization page.
You'll need admin rights in your Gong workspace to complete this authorization.
Review the permissions Zell is requesting and click **Allow** to grant access to your Gong workspace.
After authorization, a pop-up appears where you can import your Gong scorecards into Zell. Select the ones you want and click **Import**, or skip this for now.
This step is required so Zell can automatically receive your new calls from Gong.
1. In Zell, copy the **Webhook URL** shown on the Gong integration page
2. Go to your **Gong account** → **Settings** → **Admin Center**
3. Scroll down to **Ecosystem** and click **Automation Rules**
4. Click **Add a Rule**
5. Paste the webhook URL you copied from Zell
6. Give the rule a name (e.g. "Zell")
7. Click **Save**
8. **Enable the rule** — by default it's inactive. Click **Enable** so it turns on (you'll see a green check mark)
If you skip this step or forget to enable the rule, Zell won't receive your new calls automatically.
Go back to Zell and confirm that you've set up the automation rule. You'll see Gong listed as **Connected**.
You can import your last 50 calls right away. Browse the available calls on the integration page and select the ones you'd like to import.
## Set a Default Scorecard
After importing scorecards, set a **default scorecard per call type** so Zell automatically evaluates incoming calls:
1. Go to **Scorecards** in the main navigation
2. For each call type (e.g. Discovery, Demo, Follow-up), assign a default scorecard
Without a default scorecard assigned, imported calls won't be automatically evaluated.
## Importing Calls
Once connected, you can import your most recent Gong calls into Zell:
1. Go to the **Gong** integration page
2. Browse your recent calls
3. Select the calls you want to import
4. Click **Import**
Zell will pull in the full transcript, recording, and call metadata automatically.
## Importing Scorecards
You can also import your existing Gong scorecards into Zell to use them for AI-powered evaluation:
1. Go to the **Gong** integration page
2. Click **Import Scorecards**
3. Select which scorecards to import
The scorecard questions and criteria will be mapped to Zell's evaluation format.
# Manage Groups
Source: https://docs.getzell.com/guides/groups
Organize users, bots, and personas into workspace-specific groups
## Overview
Groups let you manage which users, bots, and personas belong together inside a workspace. Use them to organize teams by function, region, seniority, training program, or any other structure that determines who should work with the same Zell resources.
## How Groups Work
Each group belongs to a workspace. A group can include users, bots, and personas, and each user, bot, or persona can belong to more than one group.
Every group is created inside a workspace, so teams and training resources stay organized by workspace.
Users in a group can work with the bots and personas assigned to that group.
Users, bots, and personas can be assigned to multiple groups when responsibilities overlap.
Manage users, bots, personas, group names, and deletion from the same Groups table.
## Open Groups
Open **Settings**.
Select **Tenant Settings**.
Select **Groups**.
The Groups table shows each group's name, users, bots, personas, workspace, and actions.
## Read the Groups Table
Collapsed rows show a quick summary of each group:
* **Group Name:** The group name.
* **Users:** Up to three user avatars, plus a count for additional users.
* **Bots:** Up to three bot avatars, plus a count for additional bots.
* **Personas:** Up to three persona avatars, plus a count for additional personas.
* **Workspace:** The workspace the group belongs to.
Click the chevron at the start of a row to expand it. The expanded view shows separate sections for users, bots, and personas, with controls to add or remove assignments.
Add and remove controls are available in the expanded row. The collapsed row is for scanning the group at a glance.
## Create a Group
From the Groups page, click **Create Group**.
Choose the workspace where the group should live.
Enter a clear group name, such as "Sales Team", "Support Team", "Account Executives", "New Hires", or "Managers".
Click **Create**. The group is created immediately, and you can then expand it to assign users, bots, and personas.
Pick the workspace first. The group name field and Create button stay disabled until a workspace and group name are set.
## Rename a Group
Double-click the group name in the table.
Edit the name inline.
Press **Enter** or click the check icon to save. Press **Escape** or click the X icon to cancel.
## Assign Users
Click the chevron next to the group you want to update.
In the **Users** section, open the **Add a user** dropdown.
Search by name or email, then select the user to add them to the group.
Users already assigned to the group are not shown in the Add a user dropdown. If every available user is already in the group, the section shows **No more users available to add**.
A user can belong to multiple groups, so you do not need to remove them from one group before adding them to another.
## Assign Bots
Click the chevron next to the group you want to update.
In the **Bots** section, open the **Add a bot** dropdown.
Search for the bot, use the group filter if needed, then select the bot to add it to the group.
Bots already assigned to the group are not shown in the Add a bot dropdown. Bot rows include feature badges so you can confirm you are selecting the correct bot.
## Assign Personas
Click the chevron next to the group you want to update.
In the **Personas** section, open the **Add a persona** dropdown.
Search by persona name, job title, or industry, then select the persona to add it to the group.
Personas already assigned to the group are not shown in the Add a persona dropdown. If every persona is already assigned, the section shows **No more personas available to add**.
## Remove Assignments
To remove a user, bot, or persona from a group:
1. Expand the group.
2. Find the user, bot, or persona in its section.
3. Hover over the item.
4. Click the trash icon.
The assignment is removed from the group immediately. The user, bot, or persona itself is not deleted.
Removing a bot or persona from a group can remove access for users who only had that bot or persona through this group.
## Delete a Group
Locate the group in the table.
Hover over the row and click the trash icon in the actions column.
In the confirmation dialog, click **Delete**.
Deleting a group cannot be undone. Users, bots, and personas are not deleted, but their membership in the deleted group is removed.
## Best Practices
Use names that explain why the group exists, such as a role, program, region, or training need.
Create groups in the workspace where the users and training resources are managed.
Add people to multiple groups when they need access to resources from more than one team or program.
Regularly audit users, bots, and personas so group access stays accurate.
# Hiring Module
Source: https://docs.getzell.com/guides/hiring-module
Use Zell's hiring module to evaluate candidates through AI-powered simulations
## Overview
The Hiring Module enables you to assess job candidates using realistic AI-powered role-play scenarios before making hiring decisions. Candidates interact with AI prospects in simulated calls or situations relevant to the role, giving you objective data on their skills and approach.
## Why Use the Hiring Module?
Remove bias with standardized scenarios and AI evaluation criteria.
See how candidates handle actual situations they'll face on the job.
Screen more candidates in less time with automated simulations.
Make data-driven choices based on performance, not just interviews.
## Setting Up Hiring Assessments
### Create a Hiring Scenario
Go to **Team Management** → **Hiring** in the main menu.
Click **Create New Assessment** to begin configuring your hiring evaluation.
Enter a clear name that reflects the role and scenario:
* "Sales Rep - Cold Call Assessment"
* "Account Executive - Discovery Call Evaluation"
* "Customer Support - Difficult Customer Scenario"
Include the role name so you can easily identify assessments later.
Create assessment steps with AI bots for candidates to practice with. For each step:
* **Step Name:** Name of this practice scenario
* **Description:** Optional context about the scenario
* **Theory Content:** Optional text content or instructions for the candidate to read before the call
* **Agent:** Select which AI bot candidates will interact with
You can add multiple steps with different bots to create a comprehensive assessment.
Define the number of attempts candidates have to complete the assessment.
Allowing multiple attempts lets candidates improve and shows their learning ability.
Provide candidates with clear guidance about the assessment:
* What they'll be practicing
* What skills are being evaluated
* Any product information they can reference
* What success looks like
**Example:**
> "You'll practice calls with potential customers. Your goal is to identify their pain points, present relevant solutions, and handle objections professionally."
Choose whether candidates can see their performance results after completing
the assessment.
Click **Save** to create the hiring assessment.
## Sending Assessments to Candidates
### Send Individual Invitation
Navigate to the assessment you want to send to candidates.
Click the **Send Invitation** button to open the invitation popover.
Enter the candidate's email address in the popover.
Click **Send** to send the invitation email to the candidate.
### Invitation Validity and Expiration
Invitations have a validity period that determines how long candidates have to access and complete the assessment:
* **Default validity:** 5 days (configurable from 1-7 days when creating the assessment)
* **Expiration:** Invitations automatically expire after the validity period
Invitations cannot be accessed after they expire. If a candidate needs more
time, you'll need to send them a new invitation.
### Invitation Statuses
Invitations can have the following statuses:
* **Pending:** Invitation sent but candidate hasn't started yet
* **Started:** Candidate has begun the assessment
* **Completed:** Candidate has finished all assessment steps
* **Expired:** Invitation has passed its expiration date
* **Revoked:** Invitation was manually revoked by an admin
Candidates cannot access expired, revoked, or completed invitations. They will
see clear error messages explaining why access is blocked.
### Send Bulk Invitations
Navigate to the assessment you want to send.
Click the **Bulk Invite** button to open the bulk invitation dialog.
Enter multiple email addresses in the text area. You can separate emails by: -
Commas - Spaces - Newlines - Semicolons
Click **Send** to send invitations to all entered email addresses.
The system will show you which invitations were successful and which failed.
## Reviewing Candidate Performance
Navigate to the assessment in **Team Management** → **Hiring**.
See the invitations table showing all candidates who have been invited with:
* Candidate name
* Email address
* Status (Pending, Started, Completed, Expired, Revoked)
* Sent date
* Expiration date
* Actions (revoke, delete)
Expired invitations are automatically marked when you view the list, so you'll always see accurate status information.
See the submissions table showing all candidates who have completed or started
the assessment with: - Candidate name - Email address - Status (Pending, In
Progress, Completed) - Started date - Completed date - View action
Click **View** on any candidate submission to see their detailed performance.
On the submission detail page, you'll see a table of all their attempts
showing: - **Step Name:** Which assessment step - **Attempt Number:** Which
attempt (1, 2, 3, etc.) - **Call Duration:** Length of the call - **Status:**
Started, Submitted, Failed, or Not Started - **Completed At:** When the
attempt was completed - **Recording:** Audio player to listen to the call -
**Analysis:** Link to view detailed call analysis and evaluation
Click the analysis button (chevron icon) to view the detailed call summary, transcript, and evaluation scores for any attempt. When you open an attempt from Hiring, the page keeps Hiring navigation context so you can return to the submission or reports view.
## Managing Invitations
### Revoke an Invitation
If you need to prevent a candidate from accessing an assessment:
Navigate to the assessment in **Team Management** → **Hiring**.
Scroll to the invitations table showing all sent invitations.
Click the **Revoke** action next to the invitation you want to revoke.
Confirm the revocation. The invitation status will change to "Revoked" and the candidate will no longer be able to access it.
Revoked invitations cannot be restored. You'll need to send a new invitation
if the candidate needs access again.
### Resend an Invitation
If an invitation expires or is revoked, you can send a new one:
Navigate to the assessment.
Click **Send Invitation** and enter the candidate's email again. This creates a new invitation with a fresh expiration date.
## Privacy and Compliance
When using hiring assessments, ensure compliance with local employment laws: -
**GDPR (Europe):** Inform candidates about AI evaluation and data usage -
**EEOC (US):** Ensure assessments don't create adverse impact -
**Accessibility:** Provide accommodations for candidates with disabilities -
**Data Retention:** Follow local laws on candidate data storage
* Get explicit candidate consent for assessment and recording - Don't use
assessments as the sole hiring criteria - Validate that assessment scores
predict job success - Document your evaluation process - Provide feedback to
candidates when possible
# How to Change the Language
Source: https://docs.getzell.com/guides/language-settings
Configure language settings for your Zell account and training sessions
## Overview
Zell supports multiple languages for both the user interface and AI bot interactions. You can configure language settings to match your team's needs and regional requirements.
## Changing Your Language Settings
Click on your avatar or profile icon in the top-right corner of the application.
Select **Settings** → **Language** or **Preferences** → **Language** depending on your interface.
Choose your preferred language from the dropdown menu:
* English
* German (Deutsch)
* Italian (Italiano)
Click **Save** or **Apply** to update your language preference.
The interface will reload with your selected language.
# Connect to the Zell MCP
Source: https://docs.getzell.com/guides/mcp-integration
Set up the Zell MCP in Claude, ChatGPT, or Codex
The Zell MCP connects Claude, ChatGPT, or Codex to your Zell workspace. All clients use this remote MCP endpoint:
```text theme={null}
https://api.getzell.com/mcp
```
Authentication uses OAuth, so you do not need to create or copy an API key.
## Which client are you using?
Add Zell as a custom connector in the Claude app.
Add Zell as a custom app in ChatGPT Developer Mode.
Install Zell from your terminal and authorize it in a session.
Add and authorize Zell with the Codex CLI.
## Claude Desktop
Open the Claude desktop app and go to **Settings** → **Connectors**. Click **Add** in the upper-right corner, then select **Add custom connector**.
![Claude desktop app Connectors settings with the Add custom connector menu open and Zell shown as connected]()
An **Owner** or **Primary Owner** must first add Zell for the organization. Go to **Organization settings** → **Connectors**, click **Add**, then select **Custom** → **Web**.
Ask an **Owner** or **Primary Owner** to add Zell for your organization. After they add it, go to **Customize** → **Connectors**, find **Zell** with the **Custom** label, and click **Connect**.
Skip the next step because your organization owner has already entered the server URL.
If you are using an individual plan or adding Zell as an organization owner, use **Zell** as the connector name and enter this URL as the remote MCP server URL:
```text theme={null}
https://api.getzell.com/mcp
```
Save or connect the custom connector.
On Team and Enterprise plans, every user—including the owner who added Zell—must go to **Customize** → **Connectors**, find **Zell**, and click **Connect**.
Sign in to Zell in the browser window that opens and approve the connection.
Return to the connector list: **Settings** → **Connectors** on an individual plan or **Customize** → **Connectors** on a Team or Enterprise plan. Zell should appear as a custom web connector with a connected status.
Zell is now ready to use in Claude Desktop.
For more information about plan permissions, see [Anthropic's custom connector guide](https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp).
## ChatGPT
ChatGPT Developer Mode is available on the web for Plus, Pro, Business, Enterprise, and Education accounts. Managed workspaces may restrict who can create or publish custom apps.
Open ChatGPT on the web, go to **Settings** → **Security and login**, and turn on **Developer mode**.
Open [ChatGPT Plugins](https://chatgpt.com/plugins), click the **plus** button, and create a Developer Mode app with these details:
| Field | Value |
| -------------- | ----------------------------- |
| Name | `Zell` |
| MCP server URL | `https://api.getzell.com/mcp` |
| Authentication | OAuth |
Create the app and complete the OAuth prompt. Sign in to Zell and approve the connection. The finished app appears under **Drafts** in your ChatGPT app settings.
Start a new chat, open the **plus** menu, choose **Developer mode**, and select **Zell** for the conversation. Ask ChatGPT to use Zell in your prompt and review tool calls before approving them.
Zell is now ready to use in ChatGPT.
In a managed workspace, an admin or owner can review the draft and publish Zell for other members. Each member connects their own Zell account before first use.
For current availability and security guidance, see [OpenAI's ChatGPT Developer Mode guide](https://developers.openai.com/api/docs/guides/developer-mode).
## Claude Code
Run this command in your terminal:
```bash theme={null}
claude mcp add --transport http zell https://api.getzell.com/mcp
```
This saves Zell as an HTTP MCP server named `zell` in Claude Code.
![Terminal showing the command that adds the Zell MCP server to Claude Code]()
Start a new Claude Code session, or return to one that is already running.
Enter the following slash command inside the Claude Code session:
```text theme={null}
/mcp
```
Select **zell** and follow the authentication prompt. Claude Code opens your browser so you can sign in to Zell and authorize the connection.
![Illustrated Claude Code flow: enter slash mcp, select Zell, authorize in the browser, and return connected]()
Return to Claude Code after authorization. Open `/mcp` again and confirm that **zell** is connected.
Zell is now ready to use in Claude Code.
## Codex
Add the Zell MCP server:
```bash theme={null}
codex mcp add zell --url https://api.getzell.com/mcp
```
Then start OAuth login:
```bash theme={null}
codex mcp login zell
```
Complete the authorization in the browser window that opens. When you return to Codex, the Zell MCP server is ready to use.
## Troubleshooting
Confirm that you are using ChatGPT web with a Plus, Pro, Business, Enterprise, or Education account. In a managed workspace, ask your workspace admin whether Developer Mode or custom app creation is restricted.
Confirm that the MCP server URL is exactly `https://api.getzell.com/mcp`, refresh the Zell app from its settings, and complete the OAuth prompt again. If authorization expired, reconnect or recreate the app.
On an individual plan, add the custom connector again and confirm that the URL is exactly `https://api.getzell.com/mcp`. On Team and Enterprise plans, ask an Owner or Primary Owner to check that Zell is present under **Organization settings** → **Connectors**. After it is added there, users can find Zell under **Customize** → **Connectors**.
Run the `claude mcp add` command again and check that the server name is `zell`. Then restart your Claude Code session and open `/mcp`.
Open Zell from the Claude Desktop connector list or `/mcp` in Claude Code and start the connection again. In Codex, rerun `codex mcp login zell`. Follow any authorization URL displayed if the browser still does not open automatically.
Start the OAuth flow again from the Claude Desktop connector list, `/mcp` in Claude Code, or `codex mcp login zell` in Codex.
# Scorecards & Signals
Source: https://docs.getzell.com/guides/scorecards
Learn how to create scorecards, manage signals, and configure evaluation criteria for your team
## Overview
Scorecards evaluate calls, emails, and interactions based on **signals** — reusable evaluation criteria that can be shared across multiple scorecards. Each signal defines *what* to detect and *how* to score it.
A scorecard is simply a weighted collection of signals assigned to your team's interactions.
## Key Concepts
A single evaluation criterion with a detection method and scoring config. Signals are reusable across scorecards.
A named collection of signals, each with a weight and order. Assigned to agents or teams.
How the signal is evaluated: LLM Rubric, Keyword match, Metric, or Semantic similarity.
How much each signal contributes to the overall scorecard score (weights sum to 100%).
***
## Signal Detection Methods
Signals support four detection methods, each suited to different evaluation needs:
### LLM Rubric
The AI judges the transcript against a rubric you define. Best for nuanced, qualitative criteria.
| Field | Description |
| ------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Description** | The real-call evaluation criteria. Define what the signal checks, when it applies, and any routing, exclusion, exception, or opportunity/null rules. |
| **User Guidelines** | User-facing guidance about what good performance looks like. This also guides simulated personas, but is not used to score real calls. |
| **Rubric** | Observable scoring criteria and examples for real-call evaluation. Include any scoring-critical exceptions in the relevant bands. |
| **Mode** | `YES_NO` (binary pass/fail) or `RANGE` (0–10 scale) |
For real calls, the AI receives the signal name, Description, and Rubric. It does not receive User Guidelines. Keep every rule that could change which moments belong to the signal, whether an opportunity existed, whether the result should be null, or which score applies in the Description or Rubric.
### Keyword
Deterministic phrase matching over the transcript — no LLM needed. Fast and consistent.
| Field | Description |
| -------------------- | ------------------------------------------------------ |
| **Phrases** | List of phrases to detect (at least one required) |
| **Speaker** | Who to listen for: `rep`, `customer`, or `any` |
| **Match Word Forms** | Whether to match conjugations and plurals |
| **Bands** | Maps hit count → score (e.g., ≥1 hit → 10, 0 hits → 0) |
### Metric
A numeric measure computed from call timing data. No LLM.
| Field | Description |
| ---------- | --------------------------------------------------------------------------- |
| **Metric** | One of: `talk_ratio`, `longest_monologue`, `question_rate`, `speaking_time` |
| **Bands** | Maps measured value → score |
### Semantic
Embedding similarity match against example phrases. Useful for detecting intent without exact keyword matches.
| Field | Description |
| -------------------- | ------------------------------------------------ |
| **Prompt** | What you're looking for semantically |
| **Examples** | Optional example phrases to improve matching |
| **Threshold** | Similarity threshold (0–1) to trigger a match |
| **Pass/Fail scores** | Score when above/below threshold (default: 10/0) |
***
## Creating Signals
Go to **Team Management** → **Signals**.
Click **Create New** and choose a detection method based on your evaluation need:
* Use **LLM Rubric** for subjective or complex criteria
* Use **Keyword** for specific phrases you want to detect
* Use **Metric** for call timing measurements
* Use **Semantic** for intent-based detection
Fill in the configuration fields for your chosen method. Be specific in descriptions and provide examples where possible.
For LLM Rubric signals, make the Description and Rubric complete enough to evaluate a real call without relying on User Guidelines. Use User Guidelines to explain and practice what good execution looks like.
Signals have three statuses:
* **Published** — active and available for use in scorecards
* **Draft** — work in progress, not yet used for evaluation
* **Archived** — retired, no longer used
You cannot delete a signal that is currently in use by a scorecard. Archive it instead.
Click **Save**. The signal is now available to add to any scorecard.
***
## Creating a Scorecard
Go to **Team Management** → **Scorecards**.
Click **Create New** to start from scratch, or choose **Duplicate** to clone an existing scorecard.
Duplicating an existing scorecard is a quick way to create variations for different teams or use cases.
* Set a descriptive name (e.g., "Sales Discovery Call" or "Support Quality Check")
* Optionally add a description and difficulty level
* Select applicable call types
Add signals to your scorecard from the signal library:
* Browse or search available signals
* Click **Add** to include a signal
* Reorder signals by dragging them into the desired sequence
Each signal can only appear once per scorecard.
Assign a weight to each signal to control its contribution to the overall score. Weights are distributed as percentages that sum to 100%.
If you want equal weighting, leave the defaults — the system automatically distributes weights evenly across all signals.
Save your scorecard, then assign it to agents or teams. Run a test evaluation to verify your signals and weights produce meaningful results.
***
## Cloning & Reusing Signals
Signals are designed to be reusable:
* **Shared across scorecards** — One signal (e.g., "Proper greeting") can be used in multiple scorecards with different weights.
* **Signal library** — Global library signals (maintained by Zell) are available to all tenants. Clone them to customize.
* **Clone to edit** — Use **Clone** to create an editable copy of any signal (library or your own) that you can modify independently.
***
## Importing from Gong
When importing scorecards from Gong, each question is automatically converted into an **LLM Rubric** signal. The answer guide becomes the signal description, and all signals are assigned equal weights.
***
## Best Practices
Each signal should evaluate one specific behavior or criterion. Avoid combining multiple checks into a single signal.
Before creating a new signal, check if one already exists in the library. Reusing signals ensures consistent evaluation across teams.
LLM Rubric is powerful but slower. Use Keyword or Metric signals for simple, deterministic checks — they're faster and more predictable.
Run test evaluations after setting up a scorecard. Adjust weights and signal configs based on the results.
# Set Up Slack Bot
Source: https://docs.getzell.com/guides/slack-integration
Connect Zell to your Slack workspace for real-time notifications
## Overview
The Slack integration sends real-time notifications to your team about key events like feedback submissions, course completions, and training reminders. This guide shows you how to connect Zell to your Slack workspace.
## Setup Steps
From anywhere in the app, click your avatar and select **Tenant Settings**.
In Settings, navigate to **Integrations** (or **Apps**) and locate **Slack**.
Click **Connect to Slack**. A Slack authorization window will open for your workspace.
You'll need admin rights in your Slack workspace to complete this authorization.
1. Choose your workspace from the dropdown
2. Review the requested permissions
3. Click **Allow** to grant access
The app requires permissions to:
* Post messages to channels
* Read channel information
* Send direct messages (for personal notifications)
Back in the Zell app, you should see Slack listed as **Connected** with a green indicator.
Connection successful! Your Slack integration is now active.
## Notification Types
The integration will automatically post notifications to your configured Slack channels for these events:
When managers or peers submit feedback on a call or email
Enrollment confirmations and completion reminders
When new tasks are assigned to team members
Automated reminders for upcoming training sessions
# Assign Tasks
Source: https://docs.getzell.com/guides/tasks
Create and assign training tasks to team members with automated scheduling
## Overview
Tasks help you assign structured training activities to your team with automated scheduling and reminders. Whether it's daily practice sessions or weekly training modules, you can create recurring or one-time tasks that keep your team engaged.
## Creating and Assigning Tasks
Navigate to **Team Management** → **Tasks** in the main menu.
Click **New Task** or **+ Create Task** to open the task configuration panel.
Select from preset task types to get started quickly:
Regular training sessions
Short daily exercises
Templates are pre-configured with common settings, but you can customize everything.
Configure the fundamental task information:
* **Title:** Enter a clear, descriptive name (e.g., "Cold Call Practice with Skeptical Prospects")
* **Description:** Summarize the task goal and provide any special instructions
A good description helps team members understand why they're doing the task and what success looks like.
Choose which AI bot(s) the assignee will practice with. You can select multiple bots for variety.
Specify how many practice calls are required to complete the task (e.g., 3 calls).
Select individual team members who will receive this task.
You can assign the same task to multiple users.
Pick when the task should first appear on assignees' dashboards.
Example: If you want training to begin next Monday, set that as your start date.
Optionally set a deadline for task completion.
Determine how often the task should repeat:
One-time task that doesn't repeat.
**Best for:**
* Special projects
* Onboarding tasks
* Event-specific training
The task repeats every day.
**Best for:**
* Quick practice sessions
* Habit-building activities
* Short skill drills
The task repeats every week.
**Best for:**
* Comprehensive training sessions
* Progress reviews
* Regular skill assessments
The task repeats every month.
**Best for:**
* Monthly check-ins
* Quarterly skills assessment prep
* Longer-term practice goals
Choose the initial status:
* **Open:** Active and visible to assignees
* **Draft:** Not yet published
* **Paused:** Temporarily disabled
Add email addresses to receive reports about task completion.
Click **Create** or **Save** to enable the task.
The task will now appear on assignees' dashboards according to your schedule!
## Managing Active Tasks
### View All Tasks
Navigate to **Team Management** → **Tasks** to see:
* All active tasks
* Who's assigned to each task
* Completion rates
* Upcoming due dates
### Edit Existing Tasks
Click on any task in the list to open its details.
Update settings like:
* Title and description
* Assigned users
* Times required
* Start/end dates
* Status (open, paused, draft)
* Notification emails
Changes take effect for the task going forward.
Note: You cannot change the recurrence type after creation.
### Pause or Delete Tasks
* **Pause:** Temporarily stop a task without deleting it
* **Delete:** Permanently remove the task and all its history
Deleting a task is permanent and cannot be undone.
## Best Practices
Begin with fewer tasks and gradually increase as the team adapts to the routine.
Clear task descriptions lead to better completion rates and outcomes.
Vary task types to keep training engaging and address different skills.
Don't overload team members—balance training with their day-to-day responsibilities.
Review completion rates regularly and adjust difficulty or frequency as needed.
Recognize consistent task completion and improvement in performance.
# Connect Microsoft Teams
Source: https://docs.getzell.com/guides/teams-integration
Import Microsoft Teams recordings into Zell for automatic call analysis
## Overview
The Microsoft Teams integration lets you bring Teams meeting recordings into Zell for call analysis. After connecting Teams, Zell imports eligible recordings and adds them to your call history as real calls.
## Setup Steps
Open **Settings** and select **User Settings**.
In User Settings, go to **Integrations** and open **Call Recording**.
Click **Connect** next to **Microsoft Teams**. You'll be redirected to Microsoft to review and approve the requested permissions.
Sign in with the Microsoft account you want to connect, then approve the permissions Zell uses to read Teams meeting titles and meeting artifacts, such as recordings.
Depending on your Microsoft organization settings, an admin may need to approve the Zell permissions before the connection can be completed.
After authorization, you'll return to Zell and Teams will appear as connected.
## Imported Call Titles
When Zell imports a Teams recording, it uses the Microsoft meeting title as the call title whenever that title is available. This makes imported Teams calls easier to recognize in your call history and call analysis pages.
If Zell cannot read the meeting title, the imported call still appears in Zell using the fallback title **Teams - meeting id**.
## Existing Teams Connections
Existing Teams connections may need to be reconnected before Zell can use the updated Microsoft permissions needed to read meeting titles and meeting artifacts, such as recordings. If Teams calls continue to appear with the fallback title, disconnect and reconnect the Teams integration.
If your Microsoft organization requires admin approval, ask your Microsoft admin to approve the latest Zell permissions during reconnect.
## Disconnect Microsoft Teams
To stop importing Teams recordings, open **Settings** → **User Settings** → **Integrations** → **Call Recording** and click **Disconnect** next to **Microsoft Teams**.
Disconnecting stops Zell from using the saved Teams connection. You can reconnect later from the same integration page.
# Connect tl;dv
Source: https://docs.getzell.com/guides/tldv-integration
Import tl;dv meetings into Zell for automatic call analysis
## Overview
The tl;dv integration lets you bring customer-facing meeting recordings from tl;dv into Zell. After connecting your account, you can review recent meetings, import selected recordings manually, and optionally enable auto-import for meetings that match your filters.
## Setup Steps
From anywhere in the app, click your avatar and select **Tenant Settings**.
In Settings, go to **Integrations** and open **tl;dv**.
Paste your tl;dv API key into the **API key** field.
You can use the visibility button next to the field to show or hide the key while entering it.
Set the filters Zell should use to decide which tl;dv meetings are eligible for import:
* **Include title keywords**: only include meetings whose title contains one of these keywords.
* **Exclude title keywords**: filter out meetings whose title contains one of these keywords.
* **Minimum duration**: filter out meetings shorter than the selected number of minutes.
* **Only meetings with external participants**: filter out internal-only meetings.
* **Auto-import matching meetings**: allow Zell to import matching meetings automatically after the integration is connected.
Click **Connect to tl;dv**. Zell saves the API key and import filters, then shows the integration as connected.
## Review Recent Meetings
Once connected, the tl;dv page shows **Recent meetings**. Matching meetings are shown by default, and each row includes the meeting name, date, organizer, duration, and status.
Statuses help you decide what to import:
* **Matches**: the meeting matches your current filters and can be selected for import.
* **Filtered**: the meeting does not match one or more filters.
* **Imported**: the meeting has already been imported into Zell and cannot be selected again.
Use **Refresh** to reload the recent meetings list.
## Show Filtered Meetings
Turn on **Show filtered** to include meetings that did not match your filters. Zell shows why a meeting was filtered, such as being too short, missing an include keyword, matching an excluded title keyword, or being internal only.
Filtered meetings can still be imported manually if you decide they belong in Zell.
## Import Selected Meetings
To import meetings manually:
1. Select one or more meetings from the recent meetings list.
2. Use the header checkbox to select or clear all visible importable meetings.
3. Click **Import**.
Imported meetings are added to Zell for call analysis. After import, their status changes to **Imported**.
## Update Import Filters
After connecting tl;dv, you can update your filters from the same page. Change the filters and click **Save filters**.
Your saved filters control which meetings are shown by default, which meetings are treated as matching in the review table, and which meetings are eligible for auto-import when auto-import is enabled.
## Auto-Import
When **Auto-import matching meetings** is enabled, Zell can automatically import tl;dv meetings that match your saved filters. Meetings that are filtered out are not auto-imported, but they can still be reviewed with **Show filtered** and imported manually.
Leave auto-import turned off if you prefer to review and select meetings yourself.
## Disconnect tl;dv
To stop using the integration, open **Settings** → **Integrations** → **tl;dv** and click **Disconnect**.
Disconnecting stops Zell from using the saved tl;dv connection. You can reconnect later by entering an API key again.
# Workspaces
Source: https://docs.getzell.com/guides/workspaces
Organize your account into workspaces to separate teams, departments, or business units
## Overview
Workspaces are the top-level organizational unit in Zell. They let you separate users, bots, scorecards, personas, courses, and groups into distinct containers — ideal for multi-team or multi-department setups.
Every resource in Zell (agents, scorecards, groups, etc.) belongs to exactly one workspace, ensuring clean separation of data and access.
## Key Concepts
Each workspace contains its own users, bots, scorecards, personas, courses, and groups. Resources don't leak between workspaces.
Users within a workspace can be assigned a manager, forming a reporting hierarchy for coaching visibility.
Every account has a default workspace created automatically. New resources are placed here unless you specify otherwise.
Workspace admins can manage members and resources within their workspace without needing full account-level admin access.
***
## Managing Workspaces
### Navigate to Workspaces
Go to **Team Management** → **Workspaces**. You'll see a list of all workspaces with counts for:
* Users
* Groups
* Bots (Agents)
* Scorecards
* Personas
* Courses
***
### Create a Workspace
From the Workspaces page, click **Create Workspace**.
Choose a clear, descriptive name (e.g., "Sales EMEA", "Customer Success", "New Hire Onboarding").
Workspace names must be unique within your account.
The workspace is created and ready to have members and resources assigned.
***
### Rename a Workspace
Click on a workspace, then click **Edit**.
Change the name and save.
***
### Delete a Workspace
You must reassign all users to another workspace before deleting.
A workspace cannot be deleted while it still has users assigned. Move users first.
Click **Delete** on the empty workspace. All remaining resources (scorecards, groups, bots, etc.) within the workspace are permanently removed.
***
## Managing Members
### Assign Users to a Workspace
Navigate to the workspace you want to manage.
Use the **Members** section to select which users belong to this workspace. You can:
* Add individual users
* Bulk-assign multiple users at once
Each user belongs to exactly one workspace at a time. Assigning a user to a new workspace removes them from their previous one.
### Set a Manager
From the workspace members list, select the user you want to assign a manager to.
Set their manager from the list of available users within the same tenant.
Managers gain visibility into their direct reports' training progress, call evaluations, and coaching data.
Manager assignments cannot create cycles. You cannot assign someone as a manager if they are already a descendant in the reporting chain.
***
## What Lives in a Workspace
When you create or assign resources, they are scoped to a workspace:
| Resource | Behavior |
| ----------------- | ---------------------------------------------------- |
| **Users** | Each user belongs to one workspace |
| **Agents (Bots)** | Created within a workspace, visible only there |
| **Scorecards** | Scoped to a workspace for evaluation |
| **Groups** | Organize users and bots within the workspace |
| **Personas** | Available to users in that workspace |
| **Courses** | Training content scoped to workspace members |
| **Calls** | Logged under the workspace of the participating user |
***
## Use Cases
Separate Sales, Support, and Success teams — each with their own bots, scorecards, and training content.
Create workspaces per region (EMEA, APAC, Americas) with localized personas and evaluation criteria.
Keep new hires in an "Onboarding" workspace with beginner-level scorecards and move them to the main workspace once ramped.
For agencies or BPOs managing multiple clients, keep each client's training content and evaluations isolated.
***
## Best Practices
Don't over-segment. Start with the default workspace and only split when you need genuine isolation between teams.
If you just need to organize access to specific bots, use Groups within a workspace instead of creating separate workspaces.
Use names that make the scope obvious: team, region, or purpose.
Always move users to their new workspace before deleting the old one.
# Welcome to Zell
Source: https://docs.getzell.com/index
AI-powered training and assessment platform for sales and customer success teams
## Welcome to Zell Documentation
Zell is an AI-powered platform that helps teams practice, improve, and assess critical skills through realistic simulations and structured training programs.
Create AI-powered role-play scenarios for realistic practice sessions
Build custom scorecards to assess and track skill development
Design comprehensive courses with modules and automated progress tracking
Assess candidates through AI-powered simulations before hiring
## Key Features
Create personalized AI prospects with unique personalities, objections, and scenarios. Your team practices with realistic simulations that adapt to their responses.
[Learn how to create bots →](/guides/bots)
Define evaluation criteria that matter to your business. Track performance across categories, gather actionable insights, and identify coaching opportunities.
[Learn how to create scorecards →](/guides/scorecards)
Build multi-module training programs combining instructional content with practice sessions. Set deadlines, enable rolling enrollment, and monitor completion.
[Learn how to set up courses →](/guides/courses)
Assign recurring or one-time training tasks to individuals or groups. Automate scheduling and track completion rates.
[Learn how to assign tasks →](/guides/tasks)
Organize team members into groups for targeted training, access control, and progress tracking by department, role, or skill level.
[Learn how to manage groups →](/guides/groups)
Receive real-time notifications about feedback, completions, and key events directly in your Slack workspace.
[Learn how to set up Slack →](/guides/slack-integration)
## Quick Start Guide
Start by creating an AI prospect that simulates your ideal practice scenario.
[Create a bot →](/guides/bots)
Define what good performance looks like with custom evaluation criteria.
[Create a scorecard →](/guides/scorecards)
Set up groups and assign team members for organized training delivery.
[Manage groups →](/guides/groups)
Create tasks or courses to get your team practicing regularly.
[Assign tasks →](/guides/tasks) or [Set up a course →](/guides/courses)
## Need Help?
Explore detailed how-to guides for every feature
Get help from our team at [hey@getzell.com](mailto:hey@getzell.com)
***
**Ready to transform your team's training?** Start by exploring our [core features](/guides/bots).