The problem this solves
Most content calendars are built backwards: someone brainstorms titles in a meeting, the list gets shuffled into months, and the team writes whatever survived the discussion. Demand data enters the process late or never. The result is a quarter of publishing that feels productive while targeting queries nobody prices, clustered by gut feel instead of intent, with no view of what the current top-ranking pages already cover.
Doing it properly is genuinely tedious, which is why it rarely happens. Expanding a seed term into a full keyword universe, checking difficulty and value, clustering hundreds of terms by intent, and then reading the actual SERPs to find what the winners miss - that is days of analyst work before a single brief exists. Teams skip it, ship the brainstorm, and wonder in month four why traffic did not follow the effort.
How the mission runs
- Expand and price the keyword universe. Starting from your seed topic - revenue operations in the stock prompt - Keyword Intelligence expands into the full surrounding query space and prices it: volume, difficulty, and value signals for every term worth considering. The plan starts from the complete map rather than the ten terms everyone already knows.
- Cluster by intent. The universe gets grouped into intent clusters: what a searcher actually wants when they type each family of queries. One page per cluster, matched to where the reader is - learning, comparing, or ready to act. This is the structural decision that separates a plan from a list.
- Read the SERPs for gaps. For the priority clusters, the SERP tool pulls what currently ranks and the agent reads it critically: what the top pages cover, where they are thin, outdated, or generic. Every brief inherits a concrete angle - what to say that the current winners do not.
- Sequence the 90 days. Clusters are ordered into a quarter: quick wins early, pillar pieces where they support the clusters around them, and dependencies respected so internal links have targets. The sequence balances effort against expected impact instead of publishing in random order.
- Build the Notion board. Everything lands as a Notion database: one entry per planned piece with target cluster, priority, angle, outline, and the SERP evidence behind it. Filter by month, assign owners, and start writing - the plan arrives as a working surface rather than a slide.
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 Notion database holding your next 90 days of content: every planned piece as a card with its target keyword cluster, priced demand, intent stage, competitive angle, and a working outline. Sort it by priority or month, assign writers, and execute. The strategic thinking - expansion, pricing, clustering, gap analysis - is already done and documented on each card.
Make it yours
- Run it per product line: three parallel plans for three seed topics, each landing as its own Notion view under one database.
- Shorten the horizon: a 30-day sprint plan with only quick-win clusters, useful when you need visible movement this quarter.
- Add competitors: name two rivals and have the gap analysis weigh their coverage explicitly when choosing angles.
Frequently asked questions
How is this different from exporting a keyword list?
A list tells you what people search. This plan tells you what to build: terms clustered into pages, priced and sequenced, with a per-piece angle drawn from reading what currently ranks. The unit of output is a brief your team can write from, so the analyst work ships inside the deliverable.
Can my team edit the plan after it lands?
Yes, and they should. It arrives as a normal Notion database you own: reorder, reprioritize, kill cards, add context. The mission gives you an evidence-based starting position; editorial judgment stays with your team from the moment it lands.
What if we disagree with a cluster or an angle?
Every card carries its reasoning - the demand numbers and the SERP observations behind the angle - so disagreement is a productive review of evidence rather than a debate of opinions. Overrule anything; the next planning run learns from what you changed.