Topic Research and the Topical Map
Two tabs about what to write, from opposite directions.
- Topic Research: you go hunting for a topic
- Topical Map: you look at your site's structure and see where the holes are
Both end the same way: a brief, which becomes an article.
Topic Research
This is the pull surface. You come here with a question ("what should I write about X?") rather than waiting for the system to tell you.
Where topics come from
- A seed keyword: start from a phrase and expand outward
- Competitor gaps: topics competitors rank for and you don't
- Search Console: queries you already appear for but haven't targeted properly
- Import: bring your own list
The candidate pool
Topics you're considering collect in a pool, each carrying its real research data once research has run: keyword difficulty, monthly search volume, search intent, and whether you've already generated an article for it.
That last one matters more than it sounds. It's the check that stops you writing the same article twice, which is easy to do once you've been publishing for a few months.
Running research on a topic
Research reads the pages currently ranking for your keyword and works out what they have in common: which terms and entities recur, which facts and statistics get cited, how they're structured. That output becomes the coverage targets in your brief.
Those targets are then graded deterministically as you write. The research suggests; the checks measure. That separation is deliberate: a model proposes what to cover, but nothing model-judged ever decides your score.
The brief drawer
The brief is where research lands: your target keyword and its metrics, the terms to cover, the evidence to include, the questions to answer, internal links to weave in, and a compliance pre-check.
From there it's an article.
📸 Screenshot slot:topic-research-pool.png
The candidate pool with real KD/MSV data on several topics.
📸 Screenshot slot:brief-drawer-full.png
An expanded brief showing coverage terms, evidence and the compliance pre-check.
The Topical Map
The map is your site's actual structure, read from your published pages, not a plan you fill in.
It's a tree: pillars → silos → articles. Every node is one of two things:
- Covered: a real published page sits here
- Gap: nothing exists here yet, and something should
The map is built by reading your published content and detecting the structure that's already there: which pages are pillars, which cluster under them, and how they link to each other. Then it shows you where that structure is incomplete.
Link tracking shows whether the internal links the structure implies actually exist. A silo where nothing links to the pillar isn't really a silo. It's a pile of pages Google reads as unrelated. That's the most common structural problem on a site that grew organically, and it's invisible until something draws the tree.
The map re-scans monthly as your content grows, and you can rebuild it on demand.
📸 Screenshot slot:topical-map-tree.png
The map with a mix of covered nodes and gaps, and at least one pillar with visible link problems.
Which do I use?
Topical Map to answer "is my coverage of this subject complete, and does it hang together?" It's structural: it finds holes in what you've built.
Topic Research to answer "what's worth writing about?" It's exploratory: it finds things you hadn't considered.
And Opportunities answers the third question, "what should I do today?", by pushing you a ranked list instead of making you look.
A note on what's still landing
Both tabs are being extended. Topic Research currently works from your existing briefs plus the research and gap sources above; broader automatic discovery of net-new topic clusters is still being built. On the map, structural editing (promoting a page to a pillar, reassigning it to a different silo) is in progress, and turning a gap node directly into a brief is coming with it.
Nothing here shows you invented data. If a surface is empty, it's because there's nothing real to put in it yet.