Automations & AI reports

Advanced  Reviewing every check-in by hand doesn’t scale. An automation reads a new progress entry and drafts the analysis for you — you stay the editor and approver; the AI just does the first pass.

Automations live in Vault → Automations. Today there’s one kind: the Progress Report. The whole loop looks like this:

Set one up

From Vault → Automations → Create automation, choose the Progress Report kind, then configure:

1 · Trigger. It fires when a client submits a new progress entry. Leave it to run on any entry, or set “Trigger only on form” to scope it to one check-in form (e.g. your Weekly progress check-in). Note: an ordinary intake/survey submission doesn’t trigger it — only a progress entry does.

2 · Data fed to the AI. The heart of the analysis is this check-in plus the last few (0–5; the previous check-ins are what give the AI a trend to reason about). You can also choose to include onboarding answers, a client profile snapshot, and recent chat messages.

3 · Sections the AI fills in. A report always has the same three parts (you can reword how each is written, but not add or remove them): a client-facing summary, internal notes, and a priority flag.

4 · After drafting (post-actions). Up to five internal follow-ups — Notify trainer in app, Send email, Send WhatsApp, or Update a field on the entry (e.g. set its status). The pipeline never messages the client directly; those go out only after you approve.

5 · Save & activate. New automations save as a Draft. Hit Activate to make it live (you can Pause or Archive later). It only runs while active.

What the report contains

The automation gathers the data you chose, analyses it, and saves a draft report on that check-in with three parts:

• Client-facing summary — a short, warm message you could send the client about the week.
• Internal notes — private bullets just for you: trends, risks (low adherence, injury mentions, sharp regressions), and follow-ups for the next session.
• Priority — Standard, or Urgent when the entry suggests a red flag (a sharp regression, pain or injury, very low adherence, blood-pressure outliers, strongly negative mood/energy). Urgent reports are badged so they jump out of the list.

Each report carries a status — Draft → Approved / Discarded (or Failed if generation errored).

Review & approve

You’ll find the draft in the client’s progress view (and a “Report · Draft” chip flags the entry). Open it and you can:

• Edit the summary, the internal notes, and the priority inline.
• Approve it — quietly, or Approve & send the summary to the client. (The send button stays disabled until there’s a client-facing summary.)
• Discard a draft you don’t want, or Regenerate to get a fresh one.

The safety property that matters: nothing reaches the client until you approve it. The AI never sends on its own.

Run it & watch the runs

Once active, the automation runs itself on every matching check-in. You can also Run now… against a specific entry to test it. Each execution shows up in the automation’s run history:

Open a run to see exactly what data was sent to the AI and the raw result — handy when tuning a prompt.

Examples

• Weekly reports for everyone. Pair a Weekly Progress Tracker form (set as your check-in form) with a Progress Report automation whose trigger is “Trigger only on form → Weekly progress check-in.” Every weekly submission now drafts a report.
• Catch the urgent ones. Keep the priority section on and add a Notify trainer in app post-action — you get pinged for review, and the Urgent badge surfaces the ones that need you first.
• Auto-file the routine ones. Add an Update field on progress entry post-action to set a status (e.g. Processing) so new reports land in the right column of your Progress inbox.

Automated reports vs. asking the AI

This automation runs on every check-in, the same way each time — a dependable first draft. For open-ended questions your dashboard never anticipated (“who’s drifting?”, “compare this quarter to last”), use the conversational AI agent instead. The automation is the routine; the agent is the ad-hoc.

That’s the full loop — from a blank form to an approved report. Back to the Forms & reports overview.