Some sites rank for things they barely wrote about, because search engines have learned to associate them with a subject. Others publish one excellent article into a topic they have never touched and watch it sink. The difference is usually coverage: whether the site has demonstrably dealt with the whole subject or just one corner of it. This topical authority calculator turns that idea into a comparison you can act on — your coverage, depth, ranking performance, and internal structure, scored against the competitor currently winning.
Arb Digital's content team runs this comparison before proposing a content roadmap, because it answers the question clients actually ask. Not "is our content good", but "how much more do we need, and where".
What This Topical Authority Calculator Does
It scores four things that together describe how thoroughly a site has covered a topic. Coverage is the share of the topic's subtopics you have addressed at all. Depth is how many pages you have per covered subtopic, because one paragraph and a dedicated guide are not the same treatment. Ranking performance is how many of your pages actually reach the top ten for topic queries — the outside world's verdict on the first two. Internal structure is the share of your topic pages connected to a hub, which is what makes a set of articles read as a body of work rather than a pile.
The same four measures are calculated for a competitor, giving a gap in points and, more usefully, a gap in subtopics. That second number is the one that goes into a content plan.
How to Use It
- Build the subtopic list first. Combine your keyword research, the competitor's site structure, the questions appearing in search results, and the section headings of the top-ranking pages. Twenty to a hundred subtopics is typical for a commercial topic.
- Count your covered subtopics honestly. A subtopic counts as covered when a page addresses it substantively — not when a sentence mentions it in passing.
- Count the competitor's coverage the same way, using their site search or a crawl of their blog and resource sections.
- Pull ranking counts for both sides from your rank tracker, limited to keywords inside this topic.
- Estimate hub linking. What share of your topic pages link to, and are linked from, a central pillar page? Guessing 50% is fine; the component is weighted lightest for that reason.
The Formula: How the Score Is Built
Each component is normalised to a 0–1 index, then weighted:
Score = 100 × (0.40 × Coverage + 0.25 × Depth + 0.20 × Ranking + 0.15 × Structure)
Coverage is covered subtopics ÷ total subtopics. Depth is pages per subtopic ÷ 2, capped at 1, treating two pages per subtopic as complete treatment. Ranking is top-ten pages ÷ total subtopics, capped at 1. Structure is your hub-linking percentage.
Working the default example: coverage is 22 ÷ 60 = 0.367, depth is 1.2 ÷ 2 = 0.60, ranking is 14 ÷ 60 = 0.233, structure is 0.45. The weighted total is (0.40 × 0.367) + (0.25 × 0.60) + (0.20 × 0.233) + (0.15 × 0.45) = 0.147 + 0.150 + 0.047 + 0.068 = 0.411, so 41 out of 100. The competitor, with 48 subtopics covered, 2.1 pages each, 39 ranking pages, and 70% hub linking, scores 81 — a 39-point gap, and 26 subtopics of missing coverage.
Coverage carries the heaviest weight deliberately. It is the component that most reliably separates a site treated as a resource on a subject from one treated as an occasional visitor to it.
Where the Subtopic List Actually Comes From
The score is only as good as the denominator, and picking it badly ruins everything downstream. Too narrow a list makes you look complete when you are not; too broad a list makes an achievable project look hopeless. Four sources produce a defensible list: the section headings of the pages already ranking for your main terms, the question blocks on those results pages, the navigation of the two or three sites that dominate the topic, and your own clustered keyword research.
Cluster the keywords before counting, because raw keyword lists massively overstate the number of subtopics — thirty phrasings of one question are one subtopic, not thirty. Running the list through the keyword cluster generator first usually collapses several hundred keywords into a workable set, and each resulting cluster maps neatly to one subtopic in this calculator.
Coverage Without Depth Does Not Work
The obvious way to raise a coverage score is to publish sixty short posts, one per subtopic, in a month. It reliably fails. Thin pages published at volume tend to be evaluated as exactly that, and a large set of them can drag down how the whole site is assessed rather than lifting it. Google's guidance on creating helpful, reliable, people-first content is explicit that producing content primarily to fill gaps in search coverage is the wrong starting point.
This is why depth is weighted at 0.25 and capped at two pages per subtopic. Beyond that point, extra pages on the same subtopic usually indicate fragmentation rather than thoroughness — and fragmentation is how cannibalization begins, which the keyword cannibalization checker exists to catch. The target is complete coverage at a standard you would be happy to be judged on, added at a pace your team can sustain.
Topical Authority Is a Model, Not a Metric
No search engine publishes a topical authority score, and nobody outside those companies can measure one. What exists publicly is a body of guidance about experience, expertise, authoritativeness, and trust — the concepts described at length in Google's search quality rater guidelines, which human raters use to assess result quality. Those guidelines are not a ranking algorithm, but they describe what the systems are trying to approximate.
Read this calculator in that spirit. It is a structured way to compare your treatment of a subject against someone else's, and to convert a vague sense of being behind into a countable number of missing pages. It is not a diagnosis of how any search engine perceives your site, and the score should never be presented as if it were.
Close the Gap in the Right Order
With 26 missing subtopics, sequence matters more than speed. Start with subtopics that sit closest to your commercial pages, because coverage that supports a page which earns money returns value immediately rather than at the end of the project. Then take subtopics where the competitor ranks with visibly weak content — those are the cheapest wins in the set.
Take the depth work in parallel, not afterwards. Upgrading three existing shallow pages usually beats publishing three new ones, since the existing pages already have history and links. Then fix the structure component last, once the pages exist: a hub page linking to every subtopic, with each subtopic page linking back, is a single afternoon of work that lifts the lightest-weighted component to full marks and genuinely helps readers navigate. The internal link opportunity calculator will show how many links that actually requires.
How Long This Takes
Longer than anyone wants. Coverage has to be built page by page, each page has to be indexed and evaluated, and the ranking component only responds once the earlier work has been assessed. A realistic expectation for closing a 40-point gap is measured in quarters, not weeks, and the score will move through coverage and depth long before it moves through ranking performance.
That lag is a feature of the model rather than a flaw in it. Watching coverage climb while ranking stays flat is the normal middle phase of a topic build, and knowing that in advance stops a programme being abandoned two months before it would have started working.
Arb Digital's content team maps the full subtopic space for your market, sequences the pages by commercial impact, and builds the hub structure that ties the topic together.
Content Marketing SEO ServicesCommon Mistakes to Avoid
- Counting keywords as subtopics — thirty phrasings of one question are one subtopic, and treating them separately inflates the denominator beyond use.
- Publishing thin pages to raise coverage, which tends to lower how the whole site is assessed rather than raising it.
- Comparing against the largest site in the market instead of the one directly ahead of you, which turns a plan into a fantasy.
- Leaving the new pages unlinked, so a body of work reads as a pile of unrelated posts.
- Reporting the score as a search engine metric — it is a planning model built from your own inputs.
Related Free Tools From Arb Digital
Build the subtopic list with the keyword cluster generator and expand it with the LSI keyword generator. Check that new pages are substantial enough with the SEO content length checker, connect them using the internal link opportunity calculator, and forecast the payoff with the SEO traffic forecast calculator. The full free online tools hub has everything else.
Frequently Asked Questions
It is the idea that a site covering a subject thoroughly and consistently is treated as a stronger candidate for queries within that subject. It describes an observed pattern rather than a published metric, so any score for it is a model.
No search engine publishes a topical authority score or confirms one by that name. Public guidance instead describes experience, expertise, authoritativeness, and trust as qualities that quality assessments look for.
Between twenty and a hundred is typical for a commercial topic. Build it from competitor site structures, the headings of ranking pages, question blocks, and clustered keyword research rather than from a raw keyword export.
It raises the coverage component and usually damages everything else. Thin pages published at volume tend to be assessed as thin, and they create competing pages that fragment your own coverage.
Because it is the component that best distinguishes a site treated as a resource on a subject from one that has written about it occasionally. Depth, ranking, and structure all matter, but they build on coverage rather than replacing it.
The one directly ahead of you for your main topic queries, not the largest site in the market. A reachable target produces a plan your team can actually complete.
Usually several quarters. Coverage and depth move first as pages are published, while the ranking component responds later, once the new work has been crawled and assessed.
This score is a planning model built entirely from the figures you supply — it is not produced or endorsed by any search engine and should not be reported as one.
