Draft to publish
Authoring is where you build what a cohort sees each week: the events, the "what matters" briefings, the to-dos, and the check-in. Everything you write moves through the same simple lifecycle, which means once you have learned to publish one kind of item you have learned them all.
The status flow
A piece of content is always in one of four states, shown by the colour along its left edge:
- Draft: written, not visible to students. Warm tint.
- Scheduled: set to go live at a chosen time. Blue.
- Published: live to every matched student now. Green.
- Archived: retired, but recoverable from the archived filter. Grey.
The colours exist so you can read the state of a whole week at a glance without opening each item. A week that is all green is live; a wall of warm tint is still in progress. Archived is recoverable on purpose: retiring a piece is reversible, so you can pull something down without losing the work if you need it back.
An item can be Published and still not be on a student's screen, if it also carries a publish-at time that has not arrived yet. Published is the status you set; live is whether a student can see it right now. The counts on your home page and at the top of Briefings measure the second one, so "12 of 15 live" means twelve items a student could open today, not twelve items you have pressed Publish on. The two only differ when scheduling is involved.
You can publish several drafts at once with multi-select, which is handy when a whole week is ready to go.
The six-week arc
Content is organised by week. A cohort runs across a set number of weeks (six is typical) and each piece is scoped to a week number. The timeline view shows every week as a tab, marked as now, coming up, or complete, so you always know what is live and what is ahead.
Scoping content to a week is what makes the arc work: a briefing written for week three only surfaces in week three, so you can build the whole intake in advance and trust it to reveal itself on schedule rather than all at once.
Scheduling
Every item can carry a publish-at time. Leave it empty and the item goes live the moment you publish. Set it, and the item stays scheduled until that moment arrives, then publishes itself.
- Draft the item
Write it and leave it as a draft. Nothing is visible to students yet.
- Set a publish-at time (optional)
Choose when it should go live. Leave it empty to publish the instant you hit publish.
- Publish, or let it publish itself
A scheduled item flips to live on its own when the time arrives. You do not need to be in the console for it to happen.
A common rhythm: draft the whole week on Friday, set every item to publish Monday at 9am, and walk away. The cohort wakes up to a complete week without you touching the console over the weekend.
Audience targeting
Each item can be tagged by campus, program, and role. The facets combine with AND: an item tagged a campus and a program shows only to students who match both. Leave a facet untagged and it is universal across that facet.
This is how one cohort's feed can carry both "everyone" content and content meant only for, say, the Riverside business students.
Because facets combine with AND, every tag you add makes the audience smaller, not larger. Tagging both a campus and a program means a student must match both to see the item. If something is not appearing for people you expected, an over-tagged item is the usual cause.
The reliable way to check targeting is not to reason about it but to look. Open the cohort browser and filter by the same tags: the student count, and the names behind it, tell you whether the audience is who you meant. A count of zero, or far smaller than expected, is the fastest way to catch an over-narrow tag before students do.
The AI writing assistant
You can have the assistant draft content for you. Assistant-written items are marked Written by AI and can be filtered as a group, so nothing AI-drafted goes live unreviewed by accident. The moment you edit an AI draft, it flips back to a human draft. The assistant drafts; you decide.
The "Written by AI" mark is a safety rail, not a badge of shame. It lets you do a final sweep before a week goes live and confirm every AI-drafted item has had human eyes on it. Because editing flips the mark, the filter naturally empties as you review, leaving only the items still untouched.
Read every draft before you publish it to students. The mark and the review flow answer who decides what goes live, which is always you. They do not answer whether the words are true, and that is a separate question you have to ask of every draft.
The assistant can state a date, a room, a service name, or a policy that is simply wrong, and it will do so in the same confident tone it uses when it is right. It has no way to know your census date moved, that a service was renamed, or that a building closed. Specifics are where this bites: check every date, place, name, cost, and deadline against something authoritative before it goes in front of a cohort.
This is not a reason to avoid it. A draft you correct is still faster than a blank page. It is a reason not to skim.
One exception, worth knowing so the filter does not mislead you. If you used Draft with AI to draft a whole arc, you can edit each week inside the drafter before saving it, and those weeks still arrive marked Written by AI even though you rewrote them. The mark errs towards flagging: it will never hide something a machine wrote, but it can flag something you wrote. Editing the item again here, on Weekly briefings, flips it as usual.
Troubleshooting
cannot delete an option that already has responses
The guard on check-in options: once students have answered with an option, deleting it would orphan their responses, so the platform refuses at the database level. Edit the option's wording instead, or add a replacement and stop using the old one; the recorded answers keep their meaning either way.
A scheduled item has not appeared for students
Check three things in order: the stored publish time has actually passed (open the item and read the time back; if you scheduled it from a device set to a different timezone than your institution, confirm the saved time is the moment you meant), the item's audience tags actually include the students you are checking as (a scoped tag quietly excludes everyone else), and you are looking at the same cohort the item belongs to.
I cannot edit a colleague's item
Editing someone else's content requires its audience tags to overlap your scope, which is set on the Team page. Creating new content is not tag-restricted, so you can always author your own; ask an admin to widen your scope if you genuinely co-own a colleague's material.
Common questions
If I schedule an item, do I need to be online when it publishes?
No. A scheduled item publishes itself when its publish-at time arrives, whether or not you are in the console. That is what makes batching a week and releasing on Monday work without weekend effort.
Why is a published item not showing for some students?
Almost always audience targeting. Facets combine with AND, so each tag narrows the audience. Open the item, check its campus, program, and role tags, and filter the cohort browser by those same tags to confirm the affected student is in the audience.
Can I trust an AI-drafted item to go live as written?
No. Treat it as a draft, not a final. The assistant can get things wrong, and a wrong date or a service that no longer exists reads exactly like a right one. Check every specific against something authoritative before you publish.
The workflow helps you keep track of what you have checked: assistant items are marked Written by AI and can be filtered so you can review them as a group before a week goes live, and editing one here flips it to a human draft, which is your signal it has been reviewed. Weeks you edited inside Draft with AI before saving are the one case that stays marked despite your edits, so read the mark as "check this", not as "a machine wrote every word of this".
How often does the AI actually get something wrong?
Often enough that you should assume it will rather than hope it will not. We do not publish a number, because a number would invite you to skip the check when it looked small. The honest position is that every AI surface in First Six can produce a confident, fluent, wrong sentence, and none of them can tell you which one it is. That is why the acknowledgement sits on the screen at every one of them rather than in a policy you read once.
Next steps
Related
The fastest answer is usually one question away.