> For the complete documentation index, see [llms.txt](https://help.daasity.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.daasity.com/core-concepts/ai-analyst/give-your-ai-standing-context.md).

# Give your AI standing context

A short set of notes, with a template, that gets your AI answering in your team's terms from the right view.

Your AI answers better when it knows how your team talks about the business. Standing context is a short set of notes your AI reads at the start of every conversation. Teams that keep one get answers in their own terms, from the right view, without restating the same rules each time.

## Where it lives

* **Claude:** create a Project, connect Daasity, and add your notes as a Project file or in the Project instructions.
* **ChatGPT:** use a Project or custom instructions.
* **Other tools:** most MCP-compatible tools have a place for persistent instructions or a rules file.

Keep it short. A page or two that the whole team shares works better than a long document nobody updates.

## What to include

Copy this template and fill it in. Delete any section that doesn't apply.

```
## How we work with Daasity
- Answer from Daasity views first. Name the view you used in every answer.
- If an answer comes from anywhere other than a view, say so next to the number.
- Before quoting a number, check the view's health and how recent its data is.
- If a field seems to be missing, check what the view left out and why before saying it doesn't exist.

## Our calendar
- Fiscal year starts: [month]
- Week starts on: [day]
- "Last month" means: [last complete calendar month / last 4 weeks / ...]

## Our definitions
- Revenue we report: [e.g. Total Revenue = gross sales - discounts - refunds + shipping]
- New customer means: [...]
- DTC means: [which channels or stores]
- Orders we exclude: [e.g. replacements, test orders]

## Which view for which question
- Revenue and orders by channel: [view]
- Product and SKU performance: [view]
- Retail sell-through: [view]
- Marketing spend and ROAS: [view]

## How we compute
- AOV: [e.g. Total Revenue / valid orders]
- Ratios (ROAS, CTR, conversion rate): compute after summing the parts, never by summing row-level ratios.

## Our terms
- [Team shorthand] means [...]
```

## Keep it current

Revisit the notes when your model or your reporting changes. After a refresh, it's worth asking your AI to confirm that the views named in your notes still exist and are healthy:

"Check that every view named in my standing notes exists and tell me its health."

See [Ask your AI about your views](https://help.daasity.com/core-concepts/ai-analyst/ask-your-ai-about-your-views.md) for more questions like this.
