Sitemap + Knowledge Base: A 7-Step Topic Discovery Playbook

Most teams treat topic discovery like a creative exercise. You brainstorm, collect ideas in a doc, and pass them to a writer. The result looks busy but feels brittle. Topics do not map cleanly to your sitemap, claims drift away from your Knowledge Base, and you end up editing drafts that should have been right from the start.
The fix is not more ideas or faster writing. It is an intake system that aligns your site architecture with what your product can actually support. When sitemap taxonomy and KB facts move in lockstep, discovery becomes repeatable and publishing becomes predictable. That is how daily content becomes a configuration problem instead of a coordination problem.
Key Takeaways:
- Treat topic discovery as an intake system mapped to your sitemap and KB, not an open-ended brainstorm
- Define “done” operationally: grounded claims, narrative discipline, clean structure, planned internal links, scheduled publish date
- Audit pillars, clusters, and orphans to expose coverage debt and stranded pages
- Score each topic by business relevance and KB confidence to set a 90‑day plan
- Use a seven-part angle to pre-wire narrative, then lock claims with KB excerpts
- Write briefs that enforce structure, evidence, voice, and internal links before drafting begins
- Make governance upstream changes to Brand Studio, KB, and QA rules so quality scales without manual edits
Why Topic Discovery Breaks Without Sitemap–KB Alignment
Expose the operational gap
Most teams think topic discovery is about clever ideas, but the real failure happens when topics do not match your site’s architecture or your product’s facts. Three patterns show up quickly: ungrounded claims, repeated angles, and orphan topics. Ask a simple question for each, which of these problems is caused by missing KB evidence versus a fuzzy sitemap taxonomy. That split tells you where to fix first.
If you ship 12 posts per month and 30 percent require rework because angles overreach product capabilities, you lose four posts to cleanup. The cost is not just words. It is two to three review cycles, context switching for editors, and a broken cadence that stalls your internal link graph. Track that waste so it is visible, then stop paying it.
Link discovery to a system reality, not a writing hope. Many teams learned the hard way that faster drafting did not fix drift. See why in this explainer on AI writing limits: https://oleno.ai/ai-content-writing/why-ai-writing-didnt-fix-system. And if you want the end-to-end operating model behind the fix, start here on autonomous content systems: https://oleno.ai/ai-content-writing/why-content-requires-autonomous-systems.
Define the outcome you actually want
Write one sentence your team signs: Every topic is grounded in KB facts, framed with a seven-part angle, turned into a structured brief, then published on schedule. Pin it above your backlog. It is your acceptance test for discovery.
Set scope guardrails. No keyword volume chasing. No visibility dashboards. No forecasting. This playbook is about execution quality, not analytics. Choose a 90‑day cadence now. If you need 24 posts, approve 28 to 30 topics to absorb inevitable edits without breaking the calendar. For a complete view of the operating model, study autonomous content operations here: https://oleno.ai/ai-content-writing.
Curious what this looks like in practice? Try generating 3 free test articles now: https://savvycal.com/danielhebert/oleno-demo
Audit Your Sitemap And Map KB Gaps
Inventory pillars, clusters, and orphans
Export your sitemap into a spreadsheet and tag each URL as pillar, cluster, or orphan. Add columns for current angle, last update, and internal link weight. This turns an implicit information architecture into explicit operating choices. Expect messy edges. That is the point.
For each pillar, define the minimum cluster depth you expect, usually six to ten articles. Mark missing coverage as “coverage debt” so it gets scheduled. Then perform a link path check. If an orphan sits more than two clicks from the nearest pillar, label it “stranded” and plan either a rewrite into a cluster or a consolidation. For context on how structural drift creates operational debt, see this breakdown of content operations: https://oleno.ai/ai-content-writing/content-operations-breakdown.
Map KB to sitemap nodes
Create a column that lists the KB docs that support each pillar and cluster. Use a simple coverage scale: strong, thin, or missing. Pages without KB support are risk magnets, so flag them for KB work before they ever become drafts.
Collect product nouns that repeat across pages and match them to KB names. If naming diverges, drafts will drift. Fix naming in the KB first. Highlight KB excerpts that should anchor claims in future briefs. That simple habit compresses brief creation into minutes, not hours.
Mark knowledge deficits and risks
Add a “KB confidence” score from one to five for every topic. Score based on the number of supporting docs, their freshness, and specificity. Anything at three or below must trigger KB enrichment before writing. Capture patterns in a separate KB backlog for integrations, pricing, and edge cases. Assign a resolution owner with a short template that states required claims, where they will appear, and why they matter. Tie this to downstream quality checks with a clear line to your QA gate: https://oleno.ai/blog/build-an-automated-qa-gate-9-checks-to-prevent-bad-content.
Classify Intent And Build A Prioritization Matrix
Bucket nodes by intent
Convert sitemap nodes into intent labels you can use: informational, navigational, or commercial. Pick verbs to decide. Learn maps to informational, find maps to navigational, choose maps to commercial. If a page tries to do more than one, split it into separate topics so angles stay clean and internal links stay purposeful.
Connect intent to buyer stages you actually support. Informational content helps readers understand the problem. Commercial content clarifies solution mechanics with KB-backed specifics. Navigational pages route users to pillars and conversion pages. This dual clarity makes content easier for humans and machines to interpret, as described here: https://oleno.ai/ai-content-writing/dual-discovery-seo-llm-visibility.
Score business relevance and KB confidence
Build a simple matrix that anyone can follow. Score business relevance from one to five based on proximity to core product and must-hit launches. Score KB confidence from one to five using your earlier rules. Add cadence fit as a label, weekly or monthly, so batching is straightforward.
- Scoring rules:
- Business relevance 5: core product deep dive or launch-critical
- Business relevance 3: helpful cluster that supports a core pillar
- KB confidence 5: multiple current, precise docs
- KB confidence 2: thin, stale, or ambiguous coverage
- Cadence bonus: add 1 to priority for fixed-date topics
Calculate priority as business relevance multiplied by KB confidence, then add the cadence bonus. When two topics tie, pick the one that strengthens a pillar’s internal link map. Sequence beats speculation. For more on sequence-first thinking, explore the orchestrated pipeline method: https://oleno.ai/blog/orchestrated-content-pipeline-7-step-playbook-to-automate-publishing.
Sequence a 90-day backlog
Translate scores into a 30‑60‑90 plan. In month one, shore up pillars with high relevance and high confidence. In month two, expand clusters and fix stranded orphans. In month three, tackle lower-confidence topics after KB enrichment. Batch two to three topics weekly that share the same anchor pillar so internal links stack value. Add a brief approval day and a publish window to prevent bottlenecks. A governed editorial pipeline example lives here: https://oleno.ai/blog/build-a-governed-editorial-pipeline-topic-to-publish-in-7-steps.
Apply Seven-Part Angles That Hold Up Under Scrutiny
Fill the seven-part template
Angles should be short, specific, and grounded. Use this sequence for every topic: context, gap, intent, motivation, tension, brand point of view, demand link. Write one to two sentences per step. Keep nouns concrete and tie claims to the KB snippets you already mapped. You are pre-wiring the narrative the draft will expand.
Example: Context, teams rely on ad-hoc topic ideas. Gap, no KB grounding creates rework. Intent, commercial. Motivation, reduce editing load. Tension, faster drafting did not fix drift. Brand POV, system over manual prompts. Demand link, daily governed publishing. If you cannot read the seven lines and grasp the story, tighten nouns and swap generalities for specifics. The method fits neatly into content orchestration: https://oleno.ai/ai-content-writing/shift-toward-orchestration.
Stress-test with KB grounding
Attach a KB excerpt to each angle step that makes a factual claim. If you cannot, change the claim or enrich the KB before drafting. Label high-risk claims that must appear in the brief with evidence, not assertions. Add a “must-include evidence” checklist for features, constraints, definitions, and naming, then pick a single canonical source when multiple docs overlap.
Capture narrative consistency
Map your angle to the six-part narrative you use in drafts so the story flows the same way every time. The angle defines the content and the narrative defines the order. Both should agree on the core tension and the promised outcome. A deeper dive on the six-part narrative can help reviewers align fast: https://oleno.ai/blog/six-part-narrative-framework-turn-posts-into-persuasive-content. And if you teach commercial concepts often, reinforce the structure here: https://oleno.ai/blog/commercial-teaching-narrative-6-step-playbook-to-generate-demand.
Write Production-Ready Briefs With Enforced Rules
Define H1/H2/H3 structure and metadata
Create a brief template with a clear H1 promise, short H2 section titles, and H3s that carry one idea per subsection. Include title tag, meta description, URL slug, and alt text rules. These are formatting standards, not performance tactics, and they keep drafts clean and scannable.
- Writing standards to enforce:
- Grade‑9 readability with short paragraphs and connective language
- Consistent entities, for example “Oleno,” “Topic Intelligence,” “Knowledge Base”
- TL;DR, optional FAQs, and schema candidates when relevant
For a hub view of these structure standards, see this overview: https://oleno.ai/ai-content-writing.
Mark claims that require KB evidence
Add an evidence lock section to every brief. List claims that must include KB citations, such as feature behavior, limitations, and integration steps. Paste exact KB excerpts so writers do not paraphrase themselves into trouble. Color-code risk, green for fully grounded, yellow for partial, red for needs KB work. Topics with red items do not pass approval. Include a banned language list from Brand Studio so tone is enforced before drafting begins. To connect voice rules to checks, use this guide: https://oleno.ai/blog/build-a-brand-voice-linter-automate-consistency-across-content.
Set internal link targets
Pre-select two to three internal targets per brief. Choose one hub pillar and one or two relevant spokes. Use short, descriptive anchors that read naturally in a sentence. Only link to pages that exist, and prefer hubs when choices are equal so your architecture stays clear.
- Add one inbound target note for each new post:
- Which pillar should link back here
- Where that link should live in the pillar
For practical patterns on link planning, start here: https://oleno.ai/blog/hub-and-spoke-internal-linking-boost-seo-llm-visibility-in-90-days.
Ready to eliminate coordination drag from your brief process? Try using an autonomous content engine for always-on publishing: https://savvycal.com/danielhebert/oleno-demo
How Oleno Operationalizes This Playbook In Your Pipeline
Queue, approve, and schedule
Use the Topic Bank to hold approved topics and reorder as necessary. Set a daily output between one and 24 posts. Oleno distributes jobs so topics move through angle building, brief creation, drafting, QA, and publish without micromanagement. You decide the queue. Oleno runs the rest on schedule.
Approve topics in weekly batches. On approval, Oleno builds seven-part angles and structured briefs automatically using your Knowledge Base and Brand Studio rules. Publishing includes body, metadata, schema, and a brand-consistent hero image. If your CMS hiccups, retry logic handles it so cadence stays intact. For an overview of how steady scheduling sustains output, see this pipeline walkthrough: https://oleno.ai/blog/autonomous-publishing-pipeline-scale-to-daily-posts-without-edits.
Governance over manual edits
Keep editing out of the critical path. Tune Brand Studio for tone and phrasing. Improve the KB to close accuracy gaps. Adjust QA thresholds so drafts must pass with a minimum score before they move forward. If a draft fails, Oleno iterates and re-tests automatically. On pass, the Enhancement layer removes AI-speak, adds TL;DR, inserts internal links, and sets metadata. This is how small upstream changes improve all future output.
This approach replaces repetitive fixes with durable rules. If you want a primer on why orchestration beats ad‑hoc edits, review this perspective on content orchestration: https://oleno.ai/ai-content-writing/shift-toward-orchestration.
Publish and observe without dashboards
Publishing logs for inputs, outputs, QA scoring, publish attempts, and retries exist so the system can run predictably. They are not analytics. Use them to ensure the pipeline operates reliably, not to judge performance. Maintain steady cadence and clean structure, your internal link network will get stronger and your operations will stay calm.
Close the loop weekly. Review what shipped, note where KB coverage felt thin, and adjust Brand Studio rules where needed. That is how you drive continuous improvement without dashboards. For a hub explanation of autonomous operations and product boundaries, start here: https://oleno.ai/ai-content-writing.
Remember the rework tax on those four posts per month. Oleno removes it by automating angles, briefs, and QA before a draft ever hits your CMS. Oleno connects your sitemap, Knowledge Base, and voice into one governed pipeline so topics become scheduled articles, not threaded tasks. Oleno’s CMS connectors publish directly with schema, media, and metadata. Oleno’s QA‑Gate enforces structure, accuracy, and narrative order so the draft you see is already passable. Teams that adopt this model shift from managing edits to managing inputs, which is the only path to consistent daily publishing.
Want to see this operating model on your site in a few minutes? Try Oleno for free: https://savvycal.com/danielhebert/oleno-demo
Conclusion
Topic discovery breaks when ideas float above your sitemap and drift away from your Knowledge Base. The fix is an intake system, not a bigger brainstorm. Audit your pillars, map KB coverage, score business relevance and confidence, then run every topic through a seven-part angle and a rules-driven brief. You will ship on schedule, internal links will compound, and editing will finally leave the critical path.
If you want the pipeline that makes this playbook automatic, govern your inputs and let Oleno execute the steps. Align sitemap and KB, frame angles, lock evidence, and publish predictably. That is how you turn content from a task list into an always-on system.
About Daniel Hebert
I'm the founder of Oleno, SalesMVP Lab, and yourLumira. Been working in B2B SaaS in both sales and marketing leadership for 13+ years. I specialize in building revenue engines from the ground up. Over the years, I've codified writing frameworks, which are now powering Oleno.
Frequently Asked Questions