Cases & case data
A case is the durable record for one unit of work, holding its files, extracted data, history, and generated outputs; each template field writes its value under a stable key, so renaming a field or column later never disturbs the data already saved.
A case is the instance of a workspace's blueprint. It holds the extracted values and the documents you produce. You reach for a case when you need to keep a record of your work. The full dashboard pipeline opens on Pro.
Temporary cases
Free and Lite stay template-first. Those plans save each run as a temporary case for 14 days. Temporary cases show an `Expires {date}. Upgrade to keep it.` warning. They are capped at 50 live runs per account. You view them in a read-only dashboard.
How cases are created
Most cases are created for you. The app builds one when a file ingests from Gmail, Drive, or a manual upload. You can also build many cases at once from a roster. Drop a CSV, XLSX, or DOCX file into the Open AI Assistant chat. The assistant builds one case per row, up to 200 rows per drop, from a file of up to 15 MB.
You can create a case by hand. Click Create New to open a blank draft in your browser. A case you create while viewing one of your own mailboxes is created in that mailbox. From Inbox, Archived, Trash or All Mailboxes it lands in the Inbox. Set a Case ID. The label says Auto-generated if left blank. Add one-off fields via Add field. Uploading a file auto-saves the draft and attaches the document.
Nothing reaches the workspace until you click Create case. Untouched drafts vanish. Edited drafts prompt "Discard new case?" when you leave. Leaving a saved case with unsaved edits prompts "Discard unsaved changes?". A hard refresh triggers the browser's "Leave site?" warning.
Inside a case
People on the case show as an avatar stack in the title block. You can click Add someone to this case when it is empty. The Case actions menu holds People on this case, Add teammate, and Add client. The add form offers Teammate or Client roles.
Below the title block, values render as an unlabelled facts list. Three sections follow this list. Documents shows files that arrived and files you produced. It is one flat list where every file is a peer row. Activity calendar and Activity show a history of notes, submissions, and generations.
Identifiers and routing
Every case carries one identifier that is unique within its workspace. You set this by giving a template field the ID type, like an Invoice Number. The identifier is how files route to the right case. When an ingested file has an identifier that already exists, the file merges into that case. An unclaimed identifier starts a new case. Two workspaces can safely reuse the same identifier value.
A case with a generated identifier wears a muted auto tag. An extraction that later finds an identifier auto-promotes the case. You can change an identifier at any time. Click Edit case in the Case actions menu. Change the Case ID and click Save changes. The new value is claimed for future file routing. An empty identifier on a saved case is rejected.
Changing the identifier to an existing value opens a collision prompt. The prompt offers to `Merge into {case identifier}` or rename the case to the next free variant.
You can fix a misrouted file from the Activity feed. The row shows a Correct file ID editor. On a mismatch, it offers Move file there or `Move to new case {identifier}`. You can also use the Documents list overflow menu. It offers Move to another case... and Start a new case from this file.
Values and data storage
Extracted, answered, and delivery values are saved on the case. Computed values and AI instructions are also saved. Delivery values come from the email subject, email sender, or Drive folder. Renaming a field keeps your data safe. Rename a field or relabel a dashboard column, and the value stays exactly where it is.
Static stamps and bindings resolve at read time. They do not save to the case. They appear only in the activity feed, the template preview, and the Data tab.
You can reference values that already exist. The Value source picker offers a From case option. This opens a Case value picker. The app matches the pull by the field name. A From case pull resolves fresh on every run. It is never saved back to the case. A field can only reference data that lives on its own case.
You can add a field on a saved case. Click Add field and type a name. The app matches the typed name against existing template values and fills that value. Add field saves right away, separately from Save changes, so it never replaces a value the case already has: if the field is already filled in, the app tells you to edit that field instead.
In Edit case, each field gets the control its template declares: a calendar for dates, a number box for numbers, a checkbox for checkboxes, and for a field with choices the same dropdown, single-choice list or checkbox list you see when you run the template. A stored value that is not one of the choices stays listed until you pick another.
Provenance tags and history
Every value wears a small provenance tag. This is the same tag you see on the template preview.
- Extracted: An eye, violet. The app read it from a source file or delivery envelope.
- User Input: A pencil, amber. A person entered it.
- Computed: A square icon, teal. Derived from an expression or AI instruction.
Editing an extracted value flips its tag to User Input. Editing pins the field so later extractions leave it alone. To refresh non-pinned values, click Extract on a Documents launcher. You can also click Re-extract in the review modal's More actions menu.
A clock icon keeps the lineage one click away. The history popover shows the previous value, the current value, who changed it, and when. It also shows the document the original was read from. A plain text caption under the value names the most recent source file. You open the document from the Documents list.
Date values keep the document's original wording behind a popover next to the history icon. Address values get an `Open {field name} in Google Maps` link.
Dashboard views and columns
A field's value type decides which dashboard views the case can appear in. Address values feed the Map view. Date values feed the Calendar view. Number values feed the Analytics view.
The dashboard table builds one column for each value a field captures or derives. A computed expression can sum a column. A field that captures more than one thing shows as more than one column. An address splits into street and city columns.
Archiving and deletion
Cases are durable. Finished cases move to the Archived mailbox. Delete case in the Case actions menu permanently deletes the case after confirmation. To keep a case in Trash for 30 days, choose Move to mailbox, then Trash. A daily cleanup job permanently deletes expired cases.