> For the complete documentation index, see [llms.txt](https://docs.kula.digital/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.kula.digital/get-started/ask.md).

# What you can ask

You ask in plain English, the way you'd ask a sharp ops manager who knows your numbers. Below are starting points grouped by the problem you're trying to solve, with a note on what each one gives you back.

You don't have to phrase them exactly like this. Ask follow-ups. Push back. The AI is reading *your* data, so the more specific you get, the better the answer.

## Keeping members (retention & at-risk)

* *"Who'd love a check-in from me this week?"*
* *"Which members haven't I seen in the last 30 days who used to come regularly?"*
* *"How is the reformer cohort tracking — are they staying with us?"*
* *"Show me members whose attendance has quietly dropped off in the last two months."*

**What you get back:** a named list (when your permission level allows names) of specific members, with how long since their last visit and what changed, so you can reach out today — not a generic "improve retention" tip.

## Your classes & schedule

* *"Which classes are quietly growing, and which could use a refresh?"*
* *"What are my busiest and emptiest time slots over the last quarter?"*
* *"Is the new instructor's class finding its people?"*
* *"Which classes should I think about cancelling based on attendance trends?"*

**What you get back:** attendance trends by class and time slot, with the direction of travel — so you can make schedule calls on evidence, not gut.

## Money & cash (revenue, plans, payments)

* *"How did revenue move over the last 12 months?"*
* *"How do memberships compare to class packs for us — which holds people longer?"*
* *"Are we going to be okay next month?"*
* *"Which members are on plans that are about to lapse?"*

**What you get back:** revenue and plan trends drawn from your own payments and accounting, reconciled — with the gaps flagged where the data can't fully answer, rather than a confident guess.

## Members & outreach

* *"Draft a warm note to the members I haven't seen since May."*
* *"Who are my most loyal members I should thank this month?"*
* *"Pull together a list of members due for a plan review."*

**What you get back:** a ready-to-send draft and the list it's based on. **It writes the message; you send it** — Kula Intelligence never emails or messages members on your behalf.

## Marketing & acquisition

* *"Where are my new members actually coming from?"*
* *"How is our website traffic trending, and is it turning into bookings?"*
* *"What did we spend to acquire members last quarter, and was it worth it?"*

**What you get back:** acquisition and traffic trends from your connected marketing sources. If a source isn't connected yet, the AI tells you what's missing rather than inventing a number.

## When the answer says "I'm not sure"

A good answer sometimes is *"I can't fully answer that yet."* If a data source isn't connected, or your history has a gap, the AI will say so and tell you what to connect or fix. That honesty is the point — see [Keeping your connectors healthy](/your-data-sources/operating.md) to close gaps.

## What's next

* [Your weekly rhythm](/get-started/your-rhythm.md) — turn these into a habit.
* [Build your own & share with your team](/skills/build-your-own.md) — save your best questions as a repeatable skill.