Sending someone six separate document links and asking them to keep track is a bad way to share a course or a client deliverable. A collection is a curated set of documents behind a single link: readers land on one page, browse the documents, open any of them, and — if you allow it — chat with several documents or the whole set at once. Collections carry your branding, can be protected with a passcode or token links, and can be published in English or French. Owner admins manage them from the dashboard, on paid plans.

The obvious uses are where a single document was never the right unit: training courses, product documentation, onboarding packets, client reports shared through one token link.

The Collections tab

Open the Collections tab in the dashboard to see every collection in your organization. Each row shows the cover thumbnail, name, web address, access type, document count, and a Status chip. That chip is a switch, not a label: one click pauses a live collection or reactivates a paused one.

Link options offers Copy Link and Open in New Tab, or, for token-based collections, Select token link. Duplicate collection copies the documents and settings, including whether it is active. Delete collection removes it after confirmation, leaving the documents in your library. Create Collection starts a new one.

Inside the collection editor

The editor's side menu is grouped into Setup (Overview), Content (Documents), Experience (Layout, Tools, Chat & navigation, Disclaimer & Explainer), Security (Access, plus Sharing once saved), and — also once saved — Learning (Conversations and Intelligence).

Saving works at two speeds: toggles save on their own with a green Saved pill, while text fields wait for Apply or Save & Close. On a new collection nothing is saved until you press Create collection.

Overview holds the basics: Collection name, a URL slug, Language as English or Français, a rich-text Description, and a Cover image that falls back to your organization's default. Generate drafts a description from the keywords Docutrain extracted during training; the draft lands in the editor rather than being published. Once saved, Add translation links the collection to a sibling in the other language, giving readers English / Français switch links.

Documents has two sides: Add documents on the left, filtered with Search documents or All categories, and Selected order on the right. Only active documents can be added — but a document you pause after adding stays in the collection, ticked and marked Paused, hidden from visitors and chat until you reactivate it. Under Selected order documents are grouped by category, with A–Z and 0–9 quick-sort buttons and a drag handle for manual placement. That order is what readers see in the sidebar.

Layout is short, because every collection uses the same documentation-style reader page — no gallery variant, no layout switch. Three toggles: Show document selector in menu, Hide email contact in footer, Show landing page link.

Tools splits into Contact form, Attachments, and Image galleries, plus a Chat assistant card when your organization has its own assistant character. The contact form emails messages to the addresses in Recipient email — the first is the To address, the rest are CC'd — and every address must confirm it is willing to receive that mail first. Until it verifies, the Contact button does not appear to readers at all. Attachments become a Downloads list on the collection home page.

Chat & navigation carries Allow chat across all documents and Show references in chat answers, plus a Custom prompt card of up to 2,000 characters that works exactly like a document's, safety-screened on each save. A checkbox beneath it — Let a document's own prompt win in single-document chat — decides precedence. Checked, a document keeps its own prompt in single-document chat while multi-document chats use the collection's. Unchecked, the collection prompt applies everywhere.

Disclaimer & Explainer lets one collection-wide version replace the per-document ones. Show a collection explainer shows a single explainer once per visitor; Show a collection disclaimer requires one acceptance for the whole collection, chat across all documents included. Declining returns the reader to the collection page with a notice.

Once saved, Conversations shows the chat logs from inside the collection with the usual moderation controls, and Intelligence gives you the full Document Intelligence toolset scoped to the collection.

What a reader experiences

A visitor who follows the link passes the access check first. A paused collection never gets that far: they land on the Start page with a Collection Offline notice.

The collection opens with a header: the name, a compact line of statistics, an Introduction column holding your description, and the cover. Below sits a toolbar, and below that the Available documents list. Each document is a row with a 16:9 cover, title, optional Created date and category, a preview of its introduction, and a Read more control when there is more.

The toolbar's main control is a Chat Mode menu with up to three choices: Pick a document scrolls to the list, Select documents turns on selection mode, and Chat across all (Chat with document when there is only one) starts a chat answered from every document at once. If collection-wide chat is off, the menu becomes a single Browse documents button. A ? beside it opens a short explainer, "Should I chat with all documents?". To the right sit Downloads, galleries, translation links, and Contact.

Chatting with one document or many

Readers make this choice themselves. Chat across all searches every document at once; one document, or a few, is often quicker and clearer. Select documents puts a checkbox on every row and shows an action bar reading Chat with selected · 3, the number tracking the selection. Ctrl+K or Cmd+K starts collection-wide chat when it is enabled.

Once a chat is open, a sidebar lists every document grouped by category. Collapse sidebar shrinks it to a rail of numbered squares in the same order, each showing its document's title on hover and a Chatting badge when it is part of the live chat. Ticking a document's checkbox changes the live chat immediately — no apply step, and the last one cannot be unticked. An All documents row selects or clears everything and reports the state ("Chatting with 4 docs"). Paused documents are left out; answers come only from active members.

Per-document tools sit in the sidebar too, nested under each document behind an Available tools: row when several are in play.

Protecting and sharing a collection

The Access section opens with the Active collection toggle, then the access type: Public for anyone with the link, Passcode protected where one passcode unlocks every document inside, or Token based where only holders of a valid token link get in. Token collections get an Access tokens panel where New token takes an optional Name, Expires and Max uses, and each token can be copied, edited, deactivated, or deleted.

Two things are worth stating plainly. If you include protected documents, the editor warns you and asks you to acknowledge it on save: anyone who can open the collection can open everything in it. And signed-in members of your organization bypass both gates entirely. Show on owner landing page decides whether the collection is featured on your landing page and gallery; readers still need the passcode or token.

After saving, Sharing gives you the Share URL with a Copy button and a QR code; for token collections, share the token links from Access instead. For how these gates interact with document-level settings, see access levels explained.