We have the keyword list and no document that answers those keywords
A keyword list and a document list are different things. Put the two side by side and count the blanks, and what needs writing comes out as a list.
CONTENT SEO
Content SEO —
building the document a search engine
can actually show
A search engine cannot show a document that does not exist. We split the questions people ask by intent, map one document to one intent so they stop cannibalising each other, and bind the topic together with internal links.
In one paragraph
Content SEO is the work of having exactly one answering document per search intent. We split the questions people actually ask into informational, navigational and transactional, and map one document to each intent. Where several documents target the same intent they cannibalise each other's ranking, so we choose a canonical document and organise the rest, connecting what remains through titles, meta descriptions and internal links.
However much you fix the technology, with no document to show the result does not change. If you are indexed but get no traffic, either there is no document answering that question or the document that exists is not answering it. We count the documents already there before adding new ones.
One intent = one documentTopic clusters · internal linksThin and duplicate pages cleared
Last verified
WHEN YOU NEED THIS
When you have these problems,
this is the work you need
These are the situations we hear repeatedly in consultations. If any of them apply, start by measuring.
We wrote several pieces on the same topic and none of them ranks
When two or more documents target the same intent, a search engine has to decide which is the canonical one first. Meanwhile links and signals split, and the state where they cannibalise each other's slot sets in.
Articles have accumulated and most get no traffic
Mostly they either overlap in content with only the titles differing, or they end without answering the question. The more such documents accumulate, the more they blur the basis on which a search engine judges the whole site.
The articles are all on a blog and our own domain is empty
When everything written so far sits on an external platform, your own domain has not one document owning that question. Here, deciding which of the existing articles becomes the canonical document on your own domain comes before filling new blanks.
WHAT WE DO
What Navirang
actually does
Written as units of work rather than abstract proposals. The scope of an engagement is set from this list.
Classifying search intent and mapping documents
Queries get split into informational, navigational and transactional, and we table which document corresponds to each intent. The cells with no corresponding document are the production list.
- Collecting queries and classifying them informational, navigational, transactional
- Specifying the document form each intent needs (explanation, comparison, service, enquiry)
- Building a one-to-one correspondence table against the site's current documents
- Fixing the cells with no corresponding document as the production list
One intent, one document
Where several documents target the same intent, one becomes canonical. The rest are merged into it, or their topics are re-divided so each owns a different intent.
- Grouping the candidate documents targeting the same intent
- Selecting the canonical document and deciding how the rest are handled
- When merging, connecting the old addresses to the canonical document so the traffic path holds
- Confirming the canonical URL assignment does not disagree with the screen
Title and meta description design
The sentences that get chosen in a results list are designed separately. The title carries the document's topic; the meta description carries the reason to click this one.
- Separating the roles of the on-screen H1 and the title tag that goes to search results
- Putting the key term toward the front and fitting the length to avoid truncation
- Writing the meta description as a reason to choose rather than a summary of the body
- Checking whether the same title and description repeat across several pages
Topic clusters and internal links
One topic gets bound into a hub document and detail documents. A connected bundle of documents is what a search engine uses to grasp the site's subject.
- Separating the roles of hub and detail documents
- Two-way connection: detail → hub and hub → detail
- Matching anchor text to the topic of the document being linked
- Finding orphan documents linked from nowhere and bringing them in
Clearing thin and duplicate pages
Documents that do not contribute get judged reinforce, merge or delete. Deletion is not the only answer, and we first check whether a replacement document exists.
- Listing documents caught by none of traffic, indexing or internal links
- Reviewing template documents that only change a place name or an option
- Recording the reinforce, merge or delete judgement with its reason
- On deletion, connecting to a replacement document to avoid a dead end
Update operations
Documents carrying figures and years get a review interval attached. Before writing something new, we look at whether there is something to fix.
- Assigning review intervals to documents containing figures, years or prices
- When content changes, editing the body and the last-verified date together
- For the same topic, updating the existing URL rather than making a new one
- Resubmitting for indexing only when the content actually changed
PROCESS
In what order
does it run
What you receive at each stage is stated alongside it. Durations are the working time Navirang controls; they are not a promise about when results appear.
- 01 1 week
Content inventory
Every URL on the site goes into one table with index status, traffic and internal link count attached. It starts from a list rather than an instinct.
Content inventory
- 02 1–2 weeks
Search intent map
Queries get classified informational, navigational and transactional, and the corresponding document is connected to each intent. The blanks and the overlaps surface together.
Intent-to-document map
- 03 1 week
Clearing judgement
Each document is judged keep, reinforce, merge or delete, and the canonical documents are chosen. Reasons are recorded so they can be revisited.
Clearing judgement table
- 04 4 weeks onward
Production and restructuring
Blank documents get produced, and existing documents get their titles, meta descriptions and internal links reattached. The standard is whether the intent was answered through to the end, not length.
New and restructured documents
- 05 Ongoing
Update operations
We check index status and traffic change and manage the documents with review intervals. The cause of a change is recorded per document.
Update history · change record
DELIVERABLES
What you
receive
We do not do work that ends in conversation. The documents below remain, and become the baseline for the next measurement.
Content inventory
Every URL with index status, traffic and internal link count attached. Every later judgement happens on top of this table.
Intent-to-document map
Queries classified informational, navigational and transactional with their corresponding documents connected. The empty cells become the production list.
Clearing judgement table
The keep, reinforce, merge or delete judgement per document and its reason, marking the canonical documents and what merges into them.
Title and meta description drafts
Titles and meta descriptions written sentence by sentence for the target documents. The whole site is reviewed together so no duplicate titles remain.
Internal link structure diagram
The connection between hub and detail documents, drawn out. Orphan documents and one-way connections show up on it.
Three search intents and the documents that answer them
- Informational
- "What is AEO" — a question wanting the concept or the method. Answered with an explanatory document.
- Navigational
- "Navirang contact" — a question trying to reach a specific brand or page. That page has to be what comes up.
- Transactional
- "AEO agency cost" — a question immediately before a choice or an action. Answered with a service or enquiry document.
- The judgement principle
- One intent, one document. Where two or more target the same intent, choose a canonical one and either merge the rest or give them a different intent.
The classification test is not the form of the question but what the person asking it intends to do next. Search with the same words and a different intent needs a different document.
HOW IT CONNECTS
How it connects
to the other work
Our work moves as one piece. SEO builds the foundation for being found by search engines, AEO raises the odds of that information being cited in an answer, structured data helps machines understand the facts, and content supplies the evidence there is to cite.
| Area | Relationship to this work |
|---|---|
| Citable content design | Rewriting the same documents into paragraph structures that complete when cut |
| Technical SEO | The technical layer that lets a search engine read and process the documents you made |
| Indexing | Confirming the organised documents were actually indexed |
| Entity SEO | Making the organisation itself recognised as one entity, rather than the documents |
| SEO or AEO first | The judgement on what comes first and how the work divides |
FAQ
Frequently asked questions
Q What does content SEO actually do?
A It equips you with an answering document per search intent and organises those documents so they do not cannibalise each other. Queries get classified informational, navigational and transactional and mapped to documents; blanks get built and overlaps get merged into a canonical document. Title and meta description design, internal link structure and update operations are in scope.
Q Does writing more articles raise rankings?
A Article count and ranking are not proportional. Make several documents targeting the same intent and links and signals split, producing a state where none of them becomes canonical. Counting which intent each existing document owns, before adding more, makes results easier to confirm.
Q We already have several similar articles. Do we have to delete them?
A Deletion is not the only option. Choosing the document that answers best as canonical and merging the rest into it, or re-dividing the topics so each document owns a different intent, come first. When deleting, the old address gets connected to a replacement document so it does not become a dead end.
Q What is the difference between content SEO and citable content design?
A The outcome they aim at. Content SEO builds a document that earns a click in search results; citable content design builds paragraphs that stay complete when AI cuts them. One document can satisfy both purposes, and we write with both standards in view from the start.
Q How long does an article have to be?
A There is no set length standard. The test is whether the document answered the intent it owns through to the end. A paragraph added to fill length after the answer was finished is closer to making a thin page longer; conversely, closing a question that needs a comparison in three sentences does not answer it.
Q Should we fix old articles or write new ones?
A For the same topic we recommend fixing the existing URL. Republish at a new URL and the links and index history the existing document accumulated stay where they are, and the two documents compete. Only split into a new document when the intent it covers has itself changed, and link from the old document to the new one.
Q Can we copy blog articles onto the website as they are?
A Copying them so they sit at both addresses is not something we recommend. With the same article at two addresses a search engine has to decide again which is canonical, and if the platform's copy is chosen your own domain's document gets pushed out as the duplicate. We recommend the full text on your own domain with a summary and a link on the blog. When we classified the top 340 Naver web documents across all 17 Korean provinces on 29 July 2026, 239 of them (70.3%) were documents on organisations' own domains — it is not that own-domain documents cannot hold a place inside Naver search.
Which questions do your existing documents own?
Send us a URL and we map the document list against search intent, showing the empty cells and the overlapping ones together.
We reply within one business day.