> ## Documentation Index
> Fetch the complete documentation index at: https://docs.exec.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Practicing on Your Own
> Prep for the call you actually have, check how you are doing, and stay on top of what is assigned.
## Video walkthrough
Everything on this page works with a plain Exec account. You see your own practice, and you can build your own practice, without asking anybody's permission.
The habit worth forming is small. Instead of opening Exec to see how you did, ask the assistant you already have open. And instead of practicing whatever happens to be assigned, build the scenario for the call you actually have coming up.
Not connected yet? It takes a couple of minutes and there is nothing to switch on in Exec first. See [Connect your AI tool](/mcp/connect).
***
## Prep for the call you actually have
The reason to connect Exec to the same assistant as everything else is that almost nothing you need before a call lives in one place. The meeting is in your calendar, what was said last time is in your recorder or your notes, and the practice is in Exec. One conversation can reach all three.
### Start with what is coming up
```text theme={null}
What calls do I have coming up?
```
Your assistant reads your calendar and gives you the list. Nothing from Exec yet, and that is the point: you are establishing what you are preparing for before you ask for any practice.
### Get caught up on the last conversation
```text theme={null}
What happened on my last call with [account], and what do I need to know
before we speak again?
```
This pulls from whatever you have connected, usually your calendar plus your call recorder or notes, and comes back with a summary of last time and what is likely to come up next.
Keep going in the same thread if something stands out. The value of the next step depends entirely on how good the context is here.
### Turn it into practice
```text theme={null}
Build me a roleplay so I can practice for this call.
```
Because the context is already in the conversation, you do not have to describe the buyer. Expect a couple of clarifying questions before anything is built, roughly:
* **Who should be in the room.** If the invite has three attendees, you can practice against all three rather than one, and you get a multi-persona scenario where each person behaves differently.
* **What the scenario should put pressure on.** Name the thing you are actually worried about, whether that is scope, pricing, or a technical objection.
* **How hard it should be.**
Answer those and it writes the scenario into Exec and hands you the link. Open it and the buyer, the industry, and the objection all match the meeting you are about to walk into, because they came from the meeting rather than from a template.
A scenario you publish through an assistant is shared with your whole workspace. To keep one to yourself, just ask to set its visibility to you.
No calendar or recorder connected? This still works, you just describe the buyer yourself: who they are, what they care about, and the objection you are worried about. `Build me a roleplay to practice [the thing I want to work on] before my call. Show me the summary before you publish it.`
***
## Check how you are doing
Worth starting in a fresh thread, so the call prep above does not colour the answer.
```text theme={null}
How have I done in my most recent roleplay sessions?
```
Your scores, your ranks, and what the coach said, in one answer rather than several page loads. What comes back is a read across the set: which scenarios you have repeated, how the scores moved, and the note that keeps reappearing.
The individual criterion is usually more useful than the overall score. A session can come back strong overall and still show one objective you only partly hit, and that specific miss is the thing to practice.
### Find the pattern, not the one bad score
```text theme={null}
What feedback keeps coming up for me?
```
One low score is noise. The same note appearing in four sessions is the thing to work on, and that pattern is hard to spot when you are reading sessions one at a time.
Ask for the good example too: `show me a session where I scored well and tell me what I did right`. Knowing what your best sounds like is more useful for repeating it than another list of mistakes.
### Work out what to practice next
```text theme={null}
Based on my most recent sessions, what skills should I work on improving?
```
Your assistant goes back into the sessions it just referenced, reads the per-criterion grades and the coach feedback behind them, and groups what it finds. You get the skills that are still developing, with the specific moments that put them there rather than a bare list.
Push back if the answer is not convincing. Ask what it is based on. A recommendation resting on two sessions is worth less than one resting on twenty.
Skill proficiency and scenario analytics are limited to admins and group admins, and that applies even when you ask only about yourself, so your assistant cannot give you your aggregated proficiency score. Everything on this page is built from your own sessions, per-criterion grades, coach feedback, and assignments, which are all yours to ask about.
Then close the loop and build against it:
```text theme={null}
Build me a roleplay that targets [the skill], and publish it just for me.
```
***
## Stay on top of what is assigned
```text theme={null}
What am I assigned, and what is still outstanding?
```
You get everything assigned to you, whether it arrived through a program, a course, or straight from your manager, with the program it belongs to, how many attempts you have made, and what you still need to do to pass.
Worth asking on a Monday, and worth asking before you tell your manager you are caught up.
When you have picked one, skip the hunting:
```text theme={null}
Link me to that roleplay so I can do it.
```
You get a direct link into Exec, so you go from "what do I owe" to being in the session without opening a single menu.
***
## What other people can see
Your own practice is yours. Through the connection you see your sessions, your feedback, and your assignments, exactly as you do when you log in to Exec.
Two things worth knowing:
* **Your manager and your admins can already see your practice in Exec**, and the connection does not change that either way. It shows each person what their Exec role allows, nothing more.
[The full detail](/mcp/what-the-ai-can-see), if you want it.
***
## Make it your own
The prompts here are starting points, not a script. The ones that end up mattering are the ones you invent for your own week, so it is worth asking for things this page does not mention and seeing how far it goes.
If something does not work the way you expect, [Troubleshooting](/mcp/troubleshooting) covers the failures people actually hit, and [support@exec.com](mailto:support@exec.com) will pick up anything it does not.
***
## Next
The full prompt library, including a section for reps, and how to check an answer
What your manager and your admins can see, and what stays yours