Topical map production is the Holistic Radar service that delivers a finished topical map: every entity-attribute pair a site must cover, classified on three axes, assigned to exactly one URL, and recorded in 15 fields per row. This page defines what the service delivers, the 15-field schema, the nine production steps across two halves, and the three classification axes. It then shows a worked example row, the five quality conditions every map must pass, the three outputs a map produces, how the map hands off into content briefs, and how it differs from a keyword map. It closes with delivery format and acceptance checks, inputs the client provides, scale from the 216-page and 400+ page builds, the four failure modes production prevents, where the service sits in the Six-Layer Pyramid, and the questions buyers ask before commissioning a map.
What topical map production delivers
Topical map production delivers a topical map with 15 fields per row, the rejected rows kept visible, and a declared tuple total.
A tuple is one entity paired with one attribute of that entity. The map lists every tuple that full coverage of the subject requires, which makes coverage a fraction that can be computed rather than an opinion. The rejected rows stay in the file, so anyone can see what was considered and why it was left out.
The map is the planning document the rest of a semantic SEO engagement runs on. Content briefs, the publication sequence, the internal link graph, and the Holistic Authority Score all read from it.
The 15-field topical map schema
The topical map schema has 15 fields: two harvested from the query network, nine that are judgments, and four that follow from those judgments.
| Field | What it records | Field type |
|---|---|---|
| Attribute | The entity-attribute pair the row covers | Derived |
| Source queries | Every query from the query network that this row answers | Harvested |
| Volume | Aggregate demand across those queries | Harvested |
| Structural class | Root, seed, or node | Judgment |
| Competitive class | Standard, rare, or unique | Judgment |
| Section | Core or outer | Judgment |
| Predicate | The informational or the transactional predicate | Judgment |
| Contextual vector | Exactly one of eight vectors | Judgment |
| Prominence verdict | Whether removing the attribute breaks the central entity | Judgment |
| Relevance verdict | Semantic distance from the central entity, judged against the business | Judgment |
| Priority rank | Publication order | Judgment |
| URL slug | The destination address | Derived |
| Parent | The seed or root this row sits beneath | Derived |
| Internal link targets | Outbound links from this URL | Derived |
| Bridge target | The core page this row forwards toward | Judgment |
Only two fields are harvested. The nine judgments are where a map either produces authority or produces a backlog. A tool can fill the harvested columns in minutes; the judgment columns need a strategist who understands how the business earns money from its central entity.
Topical map production in nine steps
Topical map production runs nine steps in two halves: steps one to three build the query network, and steps four to nine apply judgment to it.
- Harvest the raw pool. Queries are collected from search suggestions, related searches, the site’s own Search Console data, and competitor ranking data.
- Annotate every query. Each query receives its question type, its contextual vector, and its entities.
- Roll entities up by type. The most frequent entity type across the pool becomes the candidate central entity.
- Select the central entity. The candidate is confirmed against the business model, not only against frequency.
- Distill the predicate pair. The source context is reduced to two predicates, one informational and one transactional, that the whole site will serve.
- Mine and filter attributes. Every candidate attribute passes or fails the prominence, popularity, and relevance filters.
- Classify on every axis. Surviving attributes receive their structural class, section, and competitive class.
- Draw the contextual border. The line between the core section and the outer section is fixed, so link direction is decided rather than improvised.
- Assign vectors, URLs, and the link graph. Each row receives one contextual vector, one URL, its parent, its link targets, and its bridge target.
The first half scales with tooling. The second half does not, because steps four to nine are judgments about a subject and a business, and adding rows per hour makes them worse. The full reasoning behind each step is set out in how to build a topical map.
The three classification axes in topical map production
Topical map production classifies every row on three independent axes: structural class, section, and competitive class.
| Axis | Values | What the value decides |
|---|---|---|
| Structural class | Root, seed, node | URL depth and internal link direction |
| Section | Core, outer | Which predicate the page serves and which way its links run |
| Competitive class | Standard, rare, unique | Where information gain is available, and so where authority is cheapest to earn |
A root attribute is inseparable from the definition of the central entity. A seed attribute branches from a root and generates its own cluster. A node attribute is terminal. A unique attribute comes from the publisher’s own source, such as a named method, original data, or internal results, and it cannot exist anywhere else. A row can be root, outer, and rare at the same time, and the combination sets its priority.
A worked topical map row
A topical map row records one tuple with all 15 fields filled, and the example below shows one row from a semantic SEO subject.
| Field | Value |
|---|---|
| Attribute | Topical map: definition |
| Structural class | Root |
| Competitive class | Standard |
| Section | Outer |
| Predicate | Knowing (informational) |
| Contextual vector | Definitional |
| Source queries | what is a topical map, topical map meaning, topical map seo, topical map definition |
| Prominence verdict | Pass: the subject has no method without it |
| Relevance verdict | Pass: direct attribute of the central entity |
| Priority rank | Early, as a root of the outer section |
| URL slug | /blog/topical-map/ |
| Parent | Semantic SEO (root) |
| Internal link targets | How to build a topical map; what is a query network |
| Bridge target | /method/six-layer-pyramid/ |
The row is live on this site as what is a topical map, and its bridge runs to the Six-Layer Pyramid. Holistic Radar publishes its own map rows because a worked example that hides its data is not a worked example; ten further rows are published in the topical map template with a real worked example.
Five quality conditions for topical map production
Every map from topical map production passes five conditions that separate a topical map from a keyword list with different column headings.
- Built from a query network. Rows come from harvested queries with their co-occurrence and sequence recorded, because sequence data sets link direction.
- Organized by attribute. Rows name entity-attribute pairs, so twelve phrasings of one question produce one complete page instead of twelve thin ones.
- Filtration verdicts recorded. Every row carries its prominence and relevance verdicts, and rejected rows stay visible.
- One contextual vector per URL. Two vectors on one URL produce a page that answers two questions and ranks for neither.
- A finite tuple total. The map declares its total, so coverage is computable as covered tuples over total tuples.
Most documents sold as topical maps pass the first two conditions and fail the last three. Those three are invisible to a buyer who has not been told to check them, and each can be checked in about ten minutes against any delivered file.
Three outputs of topical map production
Topical map production produces three outputs: a publication order, the input fields for every content brief, and the denominator for topical coverage.
The publication order sorts rows by prominence first and popularity second. Core section root attributes and quality nodes publish before anything else, and each seed cluster is completed before the next opens, because a network judged incomplete earns less than a smaller network judged complete.
The brief fields transfer directly: the attribute, its three classifications, the contextual vector, the source query cluster, the entity list, the internal link targets, the priority, and the bridge target. The tuple total becomes the divisor in every coverage measurement afterwards, including the tuple coverage component of the Holistic Authority Score, worth 30 of its 100 points.
How topical map production hands off to content briefs
Topical map production hands off to content briefs by row: each URL in the map becomes one brief, and every brief field is traced back to a map field.
The brief adds three things the map does not hold: the question inventory in answer order, the n-gram vocabulary for headings and anchors, and the information gain element the page must contain. The provenance rule, that no brief field is filled by a writer’s guess, is set out in how to write a content brief.
Topical map production compared to keyword mapping
Topical map production differs from keyword mapping in its unit, its research input, and its ability to measure a gap.
| Dimension | Topical map production | Keyword mapping |
|---|---|---|
| Unit | Entity-attribute pair | Keyword |
| Research input | Query network with sequence and path data | Keyword tool export |
| Low-volume attributes | Kept when prominent | Dropped for lack of volume |
| Rejections | Recorded with a verdict | Not recorded |
| Coverage | Computable against a tuple total | Not computable |
| Link graph | Specified per row | Left to the writer |
The practical difference is the low-volume attribute. A keyword map cannot represent an attribute nobody searches in volume, and those attributes are where prominence and information gain concentrate.
Topical map production delivery format and acceptance
Topical map production is delivered as a structured spreadsheet or database with one row per tuple, plus a written rationale for the central entity, source context, predicate pair, and contextual border.
The client accepts the map against a short checklist before any writing starts:
- Every row has all 15 fields filled.
- Rejected rows are present, with their verdicts.
- No URL carries two contextual vectors.
- Every outer row has a bridge target in the core section.
- The tuple total is stated on the first sheet.
A spreadsheet is sufficient up to a few hundred rows. A database becomes necessary when a change to one link target can no longer be traced by hand to every row that references it.
What the client provides for topical map production
Topical map production needs four inputs from the client: Search Console access, a description of how the business earns money, the list of existing URLs, and one person who can approve the central entity and source context.
The approval step matters most. A map built around the wrong central entity is internally consistent and commercially useless, so the center is agreed before a single attribute is filtered.
Topical map production scale
Topical map production has delivered a 216-page architecture for PropxPro and a 400+ page architecture for Vital Transportation.
A commercial map resolves to several hundred tuples across a smaller number of URLs, because several tuples share one page as headings or table rows. In both builds the page count was an output of the tuple inventory rather than a target set in advance. A map sized before the query network is harvested is a quota, and quotas produce thin pages. The outcomes are recorded on the Vital Transportation and PropxPro case studies.
Four failure modes topical map production prevents
Topical map production prevents the four failure modes behind most unusable maps: volume inversion, unfiltered harvest, vector deferral, and border drift.
| Failure mode | Cause | Symptom in the delivered map |
|---|---|---|
| Volume inversion | Rows ordered by volume instead of prominence | The first ten pages are commercially generic |
| Unfiltered harvest | Nothing rejected | No rejected rows and a backlog measured in years |
| Vector deferral | Vector field empty or doubled | Two of your own pages later trade positions for one query |
| Border drift | No contextual border drawn | Core pages link outward as often as outer pages link inward |
Topical map production in the Six-Layer Pyramid
Topical map production is layer three of the Six-Layer Pyramid, and it depends on layers one and two holding first.
The map cannot assign attributes without a defined central entity and source context at layer one, and it cannot set link direction without the sequence data from the query network at layer two. When either is missing, the engagement builds it first, through query network research. The method that governs the whole build is the Six-Layer Pyramid, and the service it belongs to is semantic SEO services.
Topical map production questions
How long does topical map production take?
Topical map production takes as long as its filtration takes, because the harvest scales with tooling and the prominence verdicts do not. A map delivered faster than its filtration can be reasoned about has skipped filtration, and the visible symptom is an absence of rejected rows.
Can AI build a topical map?
AI builds the harvest and the annotation well and cannot reliably make the prominence verdict, which depends on how the business earns money from the entity. A generated map includes everything relevant and ranks it by volume, which is the volume inversion failure.
Is a topical map updated after delivery?
A topical map is re-harvested annually, or immediately when the subject gains new entities or demand shifts to new attributes. Its classifications rarely change, because prominence is a property of the subject rather than of the season.
Can topical map production be bought without writing?
Topical map production is delivered on its own when a team has its own writers, with the brief fields already populated so writing starts from decisions.
How is topical map production priced?
Topical map production is priced after the free Radar Scan, because the size of the subject’s tuple inventory sets the size of the work.