How actions are prioritized

Every action on the Action Board carries a type and, in almost every case, a severity. Severity decides order; type decides what kind of work it is and where its call to action takes you. The one exception is a custom action you created without picking a severity: it has no badge and sorts to the bottom of its column.

​

Severity levels

Four levels, in board order:

• Critical: confirmed issues with direct visibility impact. The agent reserves this for problems the data clearly supports.
• High: strong evidence, meaningful impact. Setup actions (missing sitemap, crawler logs, visitor analytics, or Search Console) are always created at High, because the rest of the analysis depends on those connections.
• Medium: worth doing, used sparingly by design so the board doesn’t drown in maybes.
• Low: useful non-urgent follow-up, the rarest level. The agent and Agent Chat never create actions at Low; it appears when the agent downgrades an existing item, or when you choose it yourself.

Rules the agent follows when assigning severity:

• Evidence only: severity comes from tool data, never speculation. The agent is instructed to skip preventive “this might become a problem” actions entirely.
• Prefer the top: when choosing what to save within its per-run cap of 5, the agent prioritizes Critical first, then High, and cuts noisy Medium and Low candidates.
• Severity is revisited: on later runs, the agent re-checks open items against fresh data and can raise or lower severity. The agent only changes actions still in Suggested or To Do; once you pull something into In Progress, it leaves the severity alone.

You are not bound by the agent’s grade. The Severity dropdown in the detail dialog changes the severity of any action at any status, and Create action lets you set one (including Low) on a custom action from the start.

​

Sorting and filtering

On the board, every column sorts by severity first (Critical, High, Medium, Low, then no severity), then by created date, newest first. The same order applies in List view. The page header holds three multi-select filters that narrow every column at once: Severity (Critical, High, Medium, Low, or All severities), Assignee, and Action type. They only appear once the project has actions. Actions without a severity stay visible whatever the severity filter, so a custom action never drops out of view.

​

Action types

Nine typed actions plus one free-form type. The labels below are the ones you see on cards and in the Action type filter:

• Content Gap: create or improve content where AI answers a question but your site barely covers it. Grounded in content gap analysis; the card’s call to action runs the Content Agent.
• Sentiment: push back on unfavorable AI framing of your brand before it takes hold. See Sentiment tracking.
• Reddit: join Reddit threads AI frequently cites where your brand currently has no voice. See Reddit and YouTube strategy.
• Outreach: get your brand included in high-authority third-party articles AI keeps citing. See Offsite mentions.
• Untracked: start monitoring pages that get AI crawler traffic but aren’t tracked yet. See AI crawler insights.
• Untapped: strengthen pages AI crawlers visit often but rarely cite or drive clicks from. See Page tracker.
• Performance: speed up slow pages so AI crawlers can reliably read and index your content. See Site Health.
• Site Health: fix structural SEO issues that confuse how AI classifies your pages. See Site Health.
• Setup: connect missing integrations (sitemap, crawler logs, visitor analytics, Search Console) so the analysis covers the complete picture. Created programmatically, not by the agent.
• Custom: free-form actions created with the Create action button or through Agent Chat when no typed action fits. The discovery agent never creates these.

​

The discovery schedule

The discovery agent runs one focus per UTC day instead of everything at once:

• Monday: Content Gap.
• Wednesday: Untracked and Untapped, which share a day because both come from the same page signals.
• Thursday: Reddit.
• Friday: Outreach, limited to 3 new suggestions.
• Saturday: Sentiment.

Tuesday and Sunday have no discovery run. New Performance and Site Health suggestions are paused for now; actions of those types you already accepted stay on the board. Setup actions sit outside the rotation and are checked at the start of every run. Two conditions pause a focus for a project. Content Gap discovery is skipped when neither a sitemap nor Search Console data exists (either one is enough to run it), and it is also skipped while the Content Agent’s automation is on, since that pipeline already turns gaps into drafts; see Content Agent overview. The agent as a whole only runs for projects with a website configured and an active or trialing subscription.

​

Why a rotating focus

One focus per day is deliberate: each focus gets a full evidence-gathering pass with its own tools, and the daily cap of 5 new suggestions per run keeps the board reviewable instead of overwhelming. The agent also checks how many open items each type already has and avoids piling onto a type you haven’t cleared.

​

What this means for you

Trust the top of the board. A Critical or High action passed an evidence bar and a dedup check before it reached you, and the agent sees how many open items each type already has when it decides what to save. If you disagree with a severity, change it, or work or dismiss the action anyway; a dismissed item stays out of the Suggested column for 30 days before the agent can suggest the same issue again, and remains visible in the Dismissed column for two weeks in case you change your mind. See Working an action to done for the status flow.