Published September 3, 2026.
A topic can sound new in a planning meeting and still compete with a page already on the site. The danger is rarely an exact duplicate title. More often, two URLs answer the same search question with slightly different wording. Writers then split useful examples between them, editors struggle to explain the distinction, and internal links alternate between competing destinations. A ten-minute collision check before drafting is cheaper than sorting out that confusion after publication.
BusinessGrowthAssistant treats this as a daily routing decision, not a quarterly content audit. The operator compares the proposed article with the live corpus, records the closest neighbors, and decides whether the idea deserves a new URL. That small habit protects the daily publishing pace without turning every brief into a research project.
Start with the reader's decision
Write the proposed article's job in one sentence. Name the reader, the decision they need to make, and the constraint that makes the decision difficult. "An article about internal links" is too loose. "Help a small marketing team decide which existing page should receive an internal link when several pages mention the same service" is specific enough to compare.
This sentence matters more than the working title. Titles change during editing, while the reader's decision usually exposes overlap. Search the article library for the main object, the action, common synonyms, and the expected outcome. Open the most similar results rather than relying on titles alone.
Compare promises, not keywords
For each neighboring page, note its headline promise, intended reader, scope, and recommended next action. Two pages may share many keywords while serving different decisions. A setup guide and an exception-handling guide can coexist. Conversely, pages with different vocabulary may collide if both tell the same person to take the same action for the same reason.
Use a simple test: if a reader landed on the older page, would the proposed article still answer a material unanswered question? If the honest answer is no, strengthen the existing page instead of opening another route. A new angle is not enough when the advice and outcome remain unchanged.
Inspect the search result shape
Search intent adds another check. Review the types of pages that currently answer the query: tutorials, service pages, definitions, tools, or comparisons. This is not a command to copy competitors. It shows what kind of task searchers appear to be completing. A proposed educational article may be misplaced if the query consistently calls for a service decision, while a service page should not be stretched into a detailed operating manual.
Record the observation in plain language. Avoid declaring intent as permanent. Result pages change, and a later review may reach a different conclusion. The log should preserve what the operator saw and why it affected routing on that day.
Choose one of four dispositions
The check should end with a decision. Publish a new route when the reader's task is distinct and the site lacks a good answer. Merge the idea into an existing article when the gap is a missing section. Reframe the brief when the seed is useful but the audience or decision is still too close. Hold the topic when the distinction depends on claims or evidence the team cannot support.
Do not create a fifth category called "publish now and reconcile later." Daily output pressure makes that option tempting, but it transfers a small planning cost into a larger maintenance problem. The editor who approves the route should be able to explain its independent purpose in one sentence.
Check neighboring families
BusinessGrowthAssistant separates Blog guidance from Research evidence. A Blog article can translate a study into an operating routine, but it should not masquerade as a new study. During the collision check, search both families. If Research already holds the evidence, the Blog brief should focus on application and link to the source. If the Blog already gives the procedure, a Research proposal needs a real method, data set, or analytical question.
This family check also prevents accidental cannibalization between service pages and editorial content. A service page explains what help is available and for whom. A Blog article teaches a bounded process. The two can support each other when their roles remain clear.
Leave a compact record
The record needs the proposed slug, reader decision, three closest routes, observed differences, disposition, and approver. Include the date because the corpus and search results will change. Keep the note short enough that the next operator will actually read it.
When a collision is avoided, record where the unused material went. It may become a new section, a FAQ answer, an internal link, or a future brief with a sharper boundary. That trace prevents another writer from rediscovering the same ambiguous idea next week.
Review the check after publication
A pre-draft judgment can still be wrong. Watch whether the new route earns impressions for the intended question, whether internal-link authors know which destination to choose, and whether editors keep repeating the distinction. Confusion is evidence that the boundary exists only in the brief, not in the reader's experience.
If two pages begin competing, revisit their roles before changing URLs. One may need a narrower headline, clearer opening, stronger cross-link, or consolidation. Preserve useful content and redirects when consolidation becomes necessary. The goal is a corpus in which each route has a defensible job.
For a related release check, use the daily blog index freshness routine. When a collision decision changes an existing page, add it to the content update trigger register so the follow-through has an owner.
Frequently asked questions
Does every idea need a collision check?
Every proposed route does. The effort can be brief for an obviously distinct question, but skipping the check makes daily publishing progressively harder to govern.
Is keyword overlap proof of cannibalization?
No. Shared terms are a prompt to inspect the reader's task, page role, and observed search behavior. The decision should rest on those factors, not a word count.
Who approves a merge instead of a new article?
The content operator recommends the disposition. The accountable editor or site owner approves changes that alter an existing page's scope, URL, or public promise.
