Ask Orgo AI: your data, answered in plain language

Type what you want to know, the way you'd ask a colleague:

• "Active members count per local center"
• "Revenue by product type this year"
• "Members who have never made a payment"
• "Pending resignation requests"
• "Membership applications grouped by status"

Suggested questions on the welcome screen help new admins discover what's possible.

Ask Orgo answers in the language you ask in, and understands your organization's vocabulary — fees, chapters, applications, resignations — across languages.

If a question is ambiguous, Ask Orgo asks a clarifying question instead of guessing, and reports a confidence level with each answer. Harder questions are automatically escalated to a more capable AI model.

You can paste or drag an image into the chat — a screenshot of an old report, for example — and ask for the same numbers from your live data. Images are used for the answer and never stored.

Results show as a compact preview in the conversation, expandable to a full paginated table of every matching row.

Ask Orgo picks a fitting visualization — bar, pie, or line — and you can toggle between chart and table, or just ask: "show it as a pie chart".

Each answer comes with a short summary that points at the headline — "Top 5 local centers by revenue — North leads with 2,854" — so you're not left reading raw rows.

Any result set exports to CSV in one click, ready for a board pack or a spreadsheet.

Narrow, reshape, or extend the last answer in plain words:

• "Only active members"
• "Add a column with their email"
• "Break it down by month"
• "Same thing, but for last year"

Contextual one-click follow-ups appear under each result — "Show as chart", "Show only top 10", "How many in total?" — for the most common next steps.

Conversations are saved per admin with automatic titles, so you can search, reopen, rename, or delete past chats. Reopening a chat shows the stored result instantly, with a refresh option to re-run it against current data.

Every answer has a Save button. Give the query a name — "Unpaid fees by chapter", "Monthly registrations" — and it lands in your organization's Database Queries library.

A saved query stores the query logic, not a snapshot of the results. Opening it re-runs the query against your current data, so last month's saved report shows this month's numbers, with CSV export built in.

Saved queries are visible to admins with financial permissions, and each query can be shared with specific additional users. Every run is logged, so you can see who ran what and when.

When a result is a list of members, one click turns it into a member segment — ready for a newsletter, an event invitation, or a follow-up campaign.

Ask "create a list of members who joined this year" and Ask Orgo proposes the segment — name, criteria, and an estimated match count — then opens it in the segment editor for you to confirm.

The assistant can only ever read data. Multiple independent layers enforce this:

• The query language it uses has no write operations at all
• Every generated query is validated before it runs
• Anything that isn't a SELECT statement is refused outright

Every query is automatically restricted to your organization's own data — the scoping is added by the platform, not left to the AI.

Ask Orgo is available to organization admins only, and it tells you what it is: "Ask Orgo is AI and can make mistakes. Please double-check database entries." Thumbs-up/down feedback on every conversation helps it improve.