The problem this solves
The quarterly SEO review has a way of becoming a data recital. Someone exports Search Console, screenshots the ranking tool, arranges the charts chronologically, and presents a quarter of movement without ever quite answering the questions that matter: what did we win, what did we lose, why, and what are we betting on next. The analysis that would answer them - clustering movements, tracing losses to causes, spotting pages cannibalizing each other - is exactly the work that gets cut when the review is assembled the night before.
The other chronic failure is that reviews end at the document. The room agrees the takeaways, everyone nods at next quarter's priorities, and then the priorities live in a doc nobody reopens until they are copied, in weaker form, into the next quarterly review. Without the bridge from conclusion to committed task, the review is a ceremony that consumes a senior afternoon each quarter and changes very little about the following one.
This playbook produces both halves: the analysis with lineage, where every claimed win and loss traces to the query data behind it, and the follow-through, where each recommended bet lands in Asana as a task with context attached. The review becomes an operating document with a work queue attached, and the quarter that follows inherits committed work instead of well-intentioned notes.
How the mission runs
- Pull the quarter from Search Console. Search Console provides the quarter's organic record against the prior period: clicks, impressions, and position by query and page. The comparison windows are pinned and stated, so every movement claim in the review is reproducible by anyone who wants to check it.
- Layer in ranking intelligence. Keyword Intelligence adds the competitive frame: where you gained and lost positions, which movements track your changes versus competitor pushes or result-page shifts, and where you rank on terms whose value changed. This is the difference between reporting movement and explaining it.
- Diagnose wins, losses, and cannibalization. The agent clusters the movement into a diagnosis: wins with their likely causes, losses with theirs, and pages competing against each other for the same queries. Cannibalization gets its own section because it is the most common quietly expensive problem, and the query-level evidence for each pair is included.
- Write the review into Google Docs. The full review lands as a Google Doc: quarter summary, wins and losses with evidence, the cannibalization map, and next quarter's recommended bets ranked by expected impact. Written as analysis rather than chart inventory, it reads in fifteen minutes and every figure traces to its source data.
- File the bets as Asana tasks. Each recommended bet becomes an Asana task carrying its context: the evidence, the affected pages or clusters, and what success would look like next quarter. The review's conclusions enter the team's actual work queue the same day the document is finished.
The prompt
This is the exact objective the agent receives. Swap the obvious placeholders for your own domain, segment or channel and run it as-is from the console, Slack, or the API.
What comes back
A quarterly SEO review in Google Docs that explains the quarter rather than reciting it: wins and losses with causes, a cannibalization map with query-level evidence, and ranked bets for next quarter, every figure traceable to the data pull behind it. Plus the bets filed as Asana tasks with context attached, so the review's conclusions are in the work queue before the review meeting even convenes.
Make it yours
- Run the same structure monthly at lighter depth for the team, keeping the quarterly as the strategic checkpoint that leadership sees.
- Add a competitor section tracking two named rivals' quarters alongside yours, turning the review into a relative performance document leadership reads differently.
- Have next quarter's review open by scoring last quarter's bets - which tasks shipped and what they moved - closing the loop that most SEO programs leave open.
Frequently asked questions
How are the recommended bets chosen?
From the evidence in the same review: losses that look recoverable, wins with room to compound, cannibalization worth resolving, and gaps where ranking data shows attainable value. Each bet states its reasoning and expected impact, so the team can debate priorities on the merits before committing the quarter.
Do tasks get filed before anyone reviews the analysis?
Your call. The default files them immediately so nothing is lost between review and backlog; the gated variation delivers the Doc first and files tasks only after you approve the bets. Teams usually gate the first quarter and let subsequent quarters flow once they trust the judgment.
Can the review be reproduced or audited later?
Yes. Date ranges are pinned, definitions stated, and every figure traces to the Search Console or ranking-data call that produced it. Two quarters from now, when someone questions a number in an old review, the answer is in the document's own lineage rather than in anyone's memory.