Rubrics & Scorecards

Every simulation in Salesably ends with a score and written feedback. The rubric is the standard behind that score. Without one, AI feedback would be generic. With a well-built rubric, the AI knows exactly which behaviors to reward, which to flag, and how to describe the difference, so two reps practicing the same scenario are judged the same way and a rep who repeats a scenario can see real improvement against a fixed bar.

Why rubrics matter

• They make scoring consistent. The same behavior earns the same score every time, regardless of who (or which AI run) is evaluating.
• They make feedback specific and actionable. Because each performance level is described in words, the AI can tell a rep precisely what to do differently next time.
• They make practice measurable. Scores against a stable rubric let you track progress over sessions and across a team.
• They encode your methodology. A rubric is where your definition of a strong discovery call, a clean objection handle, or a good coaching conversation actually lives.

What a rubric is made of

Part	What it is for
Module	Ties the rubric to one scenario type (Discovery, Messaging / Pitch, Objections, Plays, Outbound & Inbound, or Management Practice). A rubric only appears as an option for scenarios of its module.
Scale	The scoring levels (for example 0, 3, 5 or a 1 to 5 scale), each with a numeric point and a plain-language label. The scale is shared by every criterion in the rubric.
Conversation Criteria	The objective, observable parts of the conversation: questions asked, value articulated, next steps secured. Stored internally as "Hard Criteria."
Communication Skills	The interpersonal qualities that are harder to measure but matter: listening, rapport, confidence, adaptability. Stored internally as "Soft Criteria."
Narrative Description	A short, user-facing explanation, written from the customer's point of view, that reps see (and can hear) before they start. It sets expectations and demystifies how they will be graded.

For each criterion you write a short description of what performance looks like at every level of the scale. That cell-by-cell detail is what lets the AI grade consistently and explain its reasoning, so keep the language concrete and behavioral.

Note on Management Practice: In most modules the AI plays a customer the rep is selling to. In Management Practice the AI plays a salesperson or another colleague and the human is the manager, so its rubrics evaluate coaching, one-on-ones, and difficult internal conversations rather than selling.

Salesably ships with proven rubrics so you can start practicing on day one, while still letting you tailor evaluation to how your team actually sells. That balance is handled by two scopes of rubric.

Default Rubrics (global)

These are the built-in, system-wide rubrics, shown under the Default Rubrics heading on each module tab.

• They are available in every studio and act as a sensible baseline.
• Studio admins cannot edit or delete them, which keeps the baseline stable across the platform. (Only Salesably super users can edit a default rubric directly.)
• You can always duplicate a default rubric to create your own editable studio copy.

Think of default rubrics as the starting template, not the finished product.

Studio Rubrics

These are rubrics that belong to your studio, shown under the Studio Rubrics heading.

• They are created by your admins, either from scratch or by duplicating a default rubric.
• They are fully editable and deletable (as long as no scenario is using them).
• They are where you encode your own criteria, scale, and language.

How they work together

When a scenario needs a rubric, Salesably prefers a studio rubric for that module if one exists, and falls back to the default rubric otherwise. So the practical path is:

1. Start with the default rubrics that already exist for each module.
2. Duplicate the one closest to your needs.
3. Edit the copy to match your methodology, then attach it to your scenarios.

Best practice: Do not customize for its own sake. Keep using defaults where they work, and create a studio rubric only when you have a real, specific reason the default does not fit.

All of your rubrics live in one place, grouped by module, so you can see at a glance what evaluation standards exist for each scenario type.

To open the Rubrics page

1. Open the Studio Settings area from the main navigation menu.
2. In the admin left sidebar, under Module Configuration, click Rubrics.

The page is titled All Rubrics / Scorecards and lives at /workspaces/admin/data/rubrics.

What you can do here

• Switch modules using the tabs across the top. Each tab shows the rubrics for one module: Discovery, Messaging / Pitch, Objections, Plays, Outbound & Inbound, Management Practice (and Stories if that module is enabled for your studio).
• See both scopes at once: your Studio Rubrics are listed first, with the Default Rubrics below them.
• Filter to what is live. Toggle Show only rubrics in use to hide rubrics that are not attached to any scenario or assignment. Your choice is remembered between visits.
• Create a new rubric with the New Rubric button.
• Duplicate, View, or Edit any rubric from its row. (Edit only appears for rubrics you are allowed to change.)
• Set Feedback Settings for the whole studio, and, if the on-the-fly module is enabled, set on-the-fly rubric preferences.

Create a rubric when no existing one captures how you want a particular kind of conversation judged. The goal is not to list every possible behavior, but to name the handful of things that actually separate a strong rep from a weak one in that scenario, and to describe each clearly enough that the AI grades the way you would.

Before you start

Duplicating an existing rubric is usually faster than starting blank. Build from scratch only when you need a genuinely different structure.

To create a rubric

1. On the Rubrics page, click New Rubric (top right).
2. Give it a Rubric Name that says what it evaluates, for example "Enterprise Discovery, Technical Buyer."
3. Choose the Module. This controls which scenarios can use the rubric, so it must match the scenario type you are scoring.
4. Set up the Rubric Scale (see "Configuring Rubric Scales"). You must define a scale before you can add any criteria.
5. Add your Conversation Criteria and Communication Skills (see "Adding Criteria to Rubrics").
6. Write or generate the Narrative Description that reps will see before practicing.
7. Click Create Rubric.

Choosing how many levels

The scale defines the bar the AI grades against. Common shapes:

• 2 levels for a simple pass / strong split, useful for a single skill you just want reinforced.
• 3 levels (for example Needs Improvement / Average / Excellent) for most everyday coaching, where you want a clear middle.
• 5 levels when you need finer resolution, for instance to distinguish stages of skill on an advanced rubric. (Five is the maximum.)

More levels mean you have to write more descriptions and the gap between adjacent levels gets subtler, so reach for 5 only when that precision earns its keep.

Choosing criteria

Keep the list short and meaningful, roughly 3 to 7 items total. A good criterion is one a coach could point to in the transcript.

Conversation Criteria (objective, observable):

• Quality and depth of discovery questions
• Whether pain and impact were uncovered
• How clearly value was tied to the buyer's situation
• Whether a concrete next step was secured
• How an objection was acknowledged and resolved

Communication Skills (interpersonal):

• Active listening and building on what the buyer said
• Rapport and tone
• Confidence and credibility
• Adapting to the buyer's style and pace

For each criterion, write what each scale level looks like in plain, behavioral language. "Asks two or more open questions that uncover a specific business pain" grades far more reliably than "asks good questions."

Tip: Pilot a new rubric with a few engaged reps and read the resulting feedback before rolling it out widely. If the AI is scoring something differently than you would, the fix is almost always sharper level descriptions.

The scale is the shared yardstick for the whole rubric. Every criterion is graded on the same levels, so the scale is where you decide both the resolution of your scoring and the words reps and coaches will see attached to each score.

What each level needs

• Point: the numeric value the level is worth (for example 0, 3, 5, or 1 through 5). These numbers drive the score.
• Label: a short plain-language name (for example "Needs Improvement," "Meets Expectations," "Excellent"). Labels are what give a raw number meaning.

To set or change the scale

1. While creating or editing a rubric, in the Rubric Scale section click Edit Scale.
2. For each level, enter its Point and Label.
3. Use Add Level to add another level, up to a maximum of 5. Use the trash icon to remove one.
4. Click Save Scale.

Keep this in mind

• A scale is required first. You cannot add criteria, or save the rubric, until the scale has at least one fully filled-in level (both point and label).
• Changing the scale reshapes every criterion. When you save a new scale, each existing criterion is updated to match the new set of levels. Adding or removing a level adds or removes that level's description across all criteria, so revisit your criterion descriptions after a scale change.
• Be consistent across rubrics. Using the same scale shape everywhere makes scores comparable from one scenario to the next. Mixing a 0 to 5 rubric with a 1 to 5 rubric makes trend reporting harder to read.

Criteria are the actual things the AI evaluates. Each one is graded on your scale, and the description you write for each level is what tells the AI (and the rep) the difference between, say, an average objection handle and an excellent one. Vague criteria produce vague scores, so this is where the quality of your coaching is won or lost.

Two categories, and why they are split

• Conversation Criteria ("Hard Criteria") are the objective, observable parts of the call: specific questions, knowledge, and outcomes. They are the easiest to grade reliably and the easiest for a rep to act on.
• Communication Skills ("Soft Criteria") are the interpersonal qualities: listening, empathy, confidence, adaptability. They matter just as much but are subtler, so their level descriptions need extra care to stay consistent.

Splitting the two keeps feedback organized and lets a rep see at a glance whether they are losing points on what they did or on how they did it.

To add a criterion

1. In either the Conversation Criteria or Communication Skills section, type a name in the New criterion name field.
2. Click Add Criterion. The criterion appears with one description box per scale level.
3. Fill in each level's description with concrete, behavioral language that says exactly what earns that score.
4. Reorder criteria by dragging them; remove one with its delete control.

What good looks like: Describe behavior you could point to in a transcript. Compare "Listened to the customer" with "Paraphrased the buyer's stated concern before responding and asked a follow-up that built on it." The second grades reliably; the first does not.

Rubrics are living standards. As your messaging, product, or coaching priorities change, the rubric should change with them, otherwise you are scoring reps against last quarter's definition of good.

To edit a rubric

1. On the Rubrics page, find the rubric and click Edit.
2. Update the name, scale, criteria, or narrative description.
3. Changes save automatically as you work (you will see a "Saving" then "Saved" indicator). You can also click Update Rubric to save and return to the list.

What you can edit

• You can edit Studio Rubrics freely.
• Default Rubrics cannot be edited by studio admins. To change one, duplicate it first and edit the copy. (Salesably super users are the exception and can edit defaults directly.)

Heads up: If a rubric is already attached to live scenarios, editing it changes how future attempts at those scenarios are scored. That is usually what you want, but be deliberate when you sharpen or relax a criterion that reps are actively being measured on.

Duplicating is the standard way to customize. Because you cannot edit a default rubric, duplicating is how you turn a built-in baseline into something you own, and it is also the fastest way to spin up a close variant of a studio rubric you already trust.

To duplicate a rubric

1. On the Rubrics page, find the rubric and click Duplicate.
2. A copy is created in your Studio Rubrics with "(Copy)" appended to the name.
3. Open the copy with Edit and adjust it.

When to use it

• Customizing a default rubric. Duplicate it, then edit the copy to fit your methodology.
• Creating a variant. For example, the same discovery rubric with a stricter bar for senior reps, or a version tuned to a specific buyer persona.

Rename the copy to something descriptive right away so your list stays scannable as it grows.

The detail view is the full, read-only picture of a rubric: its scale, every criterion, and the description at each level. Use it to review a rubric before attaching it to scenarios, to confirm exactly how reps will be graded, or to share the standard with managers.

To view a rubric

1. On the Rubrics page, click View on any rubric.
2. The detail page shows:
   • The rubric name and its module.
   • Its scope (Studio or Default) and the size of its scale (for example "3-point scale").
   • The Narrative Description, with a Play Description button to hear it read aloud.
   • The Conversation Criteria and Communication Skills, each expandable, with every scale level and its description.
   • A usage summary showing which scenarios currently use this rubric.

Also available from this page

• Edit (for studio rubrics, or any rubric if you are a super user) and Delete (for unused studio rubrics).
• Generate Lesson to turn the rubric into study material (see "Creating Lessons from Rubrics").
• Export the rubric, either Print as PDF or Export to Word, for offline review or sharing.

A rubric tells reps how they will be judged. A lesson teaches them how to meet it. From any rubric you can generate a study guide that turns your evaluation criteria into something a rep can read and learn from before they practice, so they walk into the simulation already knowing what good looks like.

Why generate a lesson

The lesson is built directly from your rubric's criteria and level descriptions, so it stays in lockstep with how reps are actually scored. It explains each criterion, shows what strong performance looks like, calls out common mistakes, and gives practical tips, all grounded in your own standard rather than generic advice. For Management Practice rubrics the guidance is framed for the manager or coach rather than for a seller.

To generate a lesson

1. Open the rubric's View page.
2. Click Generate Lesson.
3. The AI drafts a lesson from the rubric. Generation typically takes around 20 to 30 seconds.
4. Review and edit the draft in the editor, including the lesson title.
5. Choose where it lives: pick a Section and a Resource Type.
6. Click Save Lesson Plan.

After saving

• The lesson is saved as a resource in your studio's library, linked back to its source rubric.
• You can include it in learning paths and surface it to reps as preparation.
• Regenerate it whenever the rubric changes so the study material and the scoring standard never drift apart. Generating again updates the existing lesson rather than creating a duplicate.

Tip: Generating lessons for your most-used rubrics builds a library that quietly answers the question every rep has before practicing: "What exactly are they grading me on?"

Retire a rubric when it is genuinely obsolete, so your module tabs stay clean and admins are not choosing between near-duplicate standards. Salesably protects you from breaking live scenarios in the process.

To delete a rubric

1. Open the rubric's View page (or its Edit page).
2. Click Delete, then confirm.

What you can and cannot delete

• Only Studio Rubrics can be deleted. Default Rubrics cannot be removed.
• A rubric that is in use cannot be deleted. If any scenario or story is attached to it, the delete dialog lists those items with direct links. Reassign each one to a different rubric, then return and delete.

Warning: Deletion is permanent. If you might want the rubric back, duplicate or export it first.

Rubrics define what is scored. The studio's feedback directive shapes how the resulting feedback is written and what it emphasizes, across every rubric at once. It is a single, studio-wide instruction the AI applies on top of your rubrics, useful for enforcing a coaching tone or steering attention without rewriting every rubric.

Overall Directive

This is an advanced, optional setting. It is a free-text instruction to the AI about how to deliver feedback and which aspects to emphasize or ignore. It applies to AI feedback throughout your studio.

To set it

1. On the Rubrics page, find the Feedback Settings panel near the top.
2. Enter your instruction in the Overall Directive field.
3. Click Save.

Examples

• "Focus on specific, actionable next steps rather than general praise."
• "Prioritize feedback on discovery questions and active listening."
• "Balance constructive criticism with recognition of what the rep did well."
• "For our sales motion, encourage tighter, more direct closing questions."

Keep the directive short and unambiguous. Because it affects all feedback, test it on a couple of practice runs and read the output before relying on it widely.

Fair grading depends on transparency. If a rep does not know what they are being measured on, a score feels arbitrary and the feedback lands as criticism rather than coaching. Rubric explainers close that gap by showing reps, in plain language, exactly how they will be evaluated before and during practice.

The Narrative Description

This is the human-friendly front door to a rubric. It is a short explanation, written from the customer's first-person point of view, that the rep reads (and can hear) before they start.

• It frames the situation: who the customer is, what they care about, and what a good conversation feels like from their side.
• It is shown before the practice session, with optional audio playback so reps can listen instead of read.
• You can write it by hand or click Generate Description to have the AI draft one from your criteria and scale, then refine it.

To write or generate it

1. While creating or editing the rubric, find the Narrative Description field.
2. Write it directly, or click Generate Description to draft one automatically.
3. Aim to convey: the customer's situation and mindset, what they are looking for, and what success looks like.

Example: "You are about to talk with me, a busy operations lead evaluating a few vendors. I do not have time for a pitch. I want someone who asks sharp questions, actually listens to what we are struggling with, and can connect their solution to my problem. Show me you have done your homework and respect my time."

How AI grading works

1. The AI reviews the full conversation transcript.
2. It evaluates the conversation against each criterion in the rubric.
3. It assigns a score on your defined scale for each criterion.
4. It weighs the narrative description and your studio feedback directive for context and tone.
5. It returns an overall score plus specific, per-criterion feedback on strengths and what to improve.

What the rep experiences

• Before practice: the narrative description sets the scene and expectations.
• After practice: an overall score, a breakdown by criterion, concrete feedback on each point, and, on repeat attempts, a sense of how they are progressing against the same bar.

The payoff is a practice environment that feels fair: reps always know the standard, so feedback reads as a map to improvement rather than a verdict.

A rubric only does its job once it is attached to a scenario. Attaching it is what makes that rubric the standard the AI uses to score every attempt at that scenario, so this is the step that connects all your rubric work to what reps actually feel when they practice.

Attaching a rubric to a scenario

1. When you create or edit a scenario (Discovery, Messaging / Pitch, Objections, Plays, Outbound & Inbound, or Management), the form includes a rubric selector.
2. Only rubrics for that scenario's module are offered, with your studio rubrics listed ahead of the defaults.
3. If you do not pick one, Salesably uses your studio rubric for that module when one exists, and otherwise the default rubric.

Where rubric-based scoring shows up

• Practice and scored simulations, where reps converse with the AI and get scored against the rubric.
• Learning paths and assignments that include those scenarios.
• Any place a scenario's results are reviewed, since the rubric structures the breakdown.

On-the-fly rubric preferences

If the on-the-fly module is enabled for your studio, an On-the-fly rubric preferences button appears in the Feedback Settings panel. There you can choose a default rubric to apply per conversation type when an on-the-fly scenario matches it, so spontaneous practice is still scored against a standard you chose.

Seeing where a rubric is used

Each rubric's detail page lists the scenarios attached to it, and the Show only rubrics in use filter on the main list hides rubrics that are not attached to anything. Together they make it easy to spot unused rubrics to retire and to see which standards are actually shaping your team's practice.

Rubrics are the most leveraged thing you configure in Salesably, because they shape every score and every piece of feedback. A small set of well-built rubrics will do more for your team than a large, sprawling library. Use these guidelines to keep yours sharp.

Start from defaults, customize with intent

• Review the default rubrics for each module before building anything.
• Duplicate the closest default and edit the copy rather than starting blank.
• Create a studio rubric only when you can name the specific way the default falls short.

Keep criteria few and concrete

• Aim for roughly 3 to 7 criteria. Too many and reps cannot focus, and the AI's attention is spread thin.
• Describe observable behavior at each level. "Asks an open question that uncovers business impact" beats "asks good questions" every time.
• Avoid vague, judgment-loaded words in level descriptions; they produce inconsistent scoring.

Keep scales consistent

• Use the same scale shape across rubrics so scores are comparable from scenario to scenario.
• Remember that changing a scale rewrites every criterion in the rubric, so revisit descriptions after any scale change.

Roll out deliberately

1. Pilot a new or revised rubric with a few engaged reps.
2. Read the actual AI feedback it produces and refine the level descriptions where the scoring does not match your judgment.
3. Generate a lesson from the rubric so reps can prepare.
4. Make sure managers understand the standard before it drives reviews.

Maintain over time

• Use clear, descriptive names so a growing list stays scannable.
• Periodically use the Show only rubrics in use filter to find and retire rubrics nothing is attached to.
• Revisit rubrics when your messaging, product, or process changes, so reps are always measured against your current definition of good.

Common pitfalls

Pitfall	Fix
Too many criteria	Trim to the handful that actually distinguish strong from weak.
Vague level descriptions	Rewrite in concrete, behavioral terms tied to what you would see in a transcript.
Inconsistent scales across rubrics	Standardize on one scale shape.
Set and forget	Schedule periodic reviews and update as your methodology evolves.
No narrative description	Always include one so reps know the standard before they practice.

Remember: The signal that a rubric needs work is almost always in the feedback it generates. When the AI scores something differently than you would, sharpen the level descriptions rather than adding more criteria.