What Is Topical Authority, and Is It Measurable?

What do people mean by topical authority?

The phrase describes a claim about how search engines treat subject expertise: that a site covering one subject thoroughly and consistently will, over time, rank better across that subject than a site touching it occasionally. Publish enough about a topic, the argument runs, and the engine comes to regard you as a source on it.

As an account of how to build something worth reading, this is sensible. A site that covers a field properly accumulates internal links, returning readers, citations from others and a body of work that answers the questions people actually ask. Those are real properties with real effects.

The trouble begins when the description hardens into a mechanism — a score held somewhere inside Google, accruing as content is published, that can be raised by buying coverage or commissioning articles. That version implies something measurable, purchasable and trackable.

So the question worth asking is not whether covering a subject well helps. It plainly does, for reasons that need no special mechanism. The question is whether topical authority names a thing that exists, can be measured, and can be bought — because that is what is being sold.

The phrase also travels under other names — subject authority, entity authority, topical relevance — and the substitutions matter less than the shared structure. Each proposes an accumulating property of a site, distinct from the quality of any individual page, that rises with published volume on a theme.

Does Google name any such system?

No. Google publishes a guide to the ranking systems used in Search, and it is specific. The seventeen named systems are BERT, crisis information systems, deduplication systems, the exact match domain system, freshness systems, link analysis systems and PageRank, local news systems, MUM, neural matching, original content systems, removal-based demotion systems, the passage ranking system, RankBrain, reliable information systems, the reviews system, the site diversity system, and spam detection systems. Four more are listed as retired, including Panda and Penguin.

There is no topical authority system on that list, under that name or an equivalent one. Nor is there a subject-expertise score, a domain specialisation classifier, or anything of the sort.

This mirrors the position with domain authority, examined separately in this series, and the two errors have the same shape: a plausible intuition about how search works, given a name, then treated as a quantity that exists and can be moved.

The honest qualification is the same as well. Google does not publish a complete account of its internals, so absence from the documentation is evidence about what Google has described rather than proof that no subject-related signal exists. What it does establish is that anyone quoting a topical authority score, or a threshold for achieving one, is not citing Google.

What does Google describe instead?

Something more modest and more useful. The guide states: “Our ranking systems are designed to work on the page level, using a variety of signals and systems to understand how to rank individual pages. Site-wide signals and classifiers are also used and contribute to our understanding of pages.”

Site-level information therefore exists and contributes. What it does not do is decide outcomes wholesale, and Google says so directly: “Having some good site-wide signals does not mean that all content from a site will always rank highly, just as having some poor site-wide signals does not mean all the content from a site will rank poorly.”

Read carefully, that is close to the opposite of how topical authority is usually sold. The selling version says a site accumulates standing that lifts everything it publishes. Google's version says ranking is decided page by page, with site-level information contributing to the understanding of individual pages and guaranteeing nothing about any of them.

Both statements can describe a site that covers a subject well ranking better across it. The difference is that Google's account offers nothing to purchase, no threshold to cross, and no number to report.

Why is the concept so appealing?

Because it converts an unbounded problem into a plan. Being a source worth citing is difficult and slow; publishing thirty articles on a subject cluster is a schedule, a budget and a deliverable.

It also provides a rationale for volume, which suits sellers of volume. If authority accrues with coverage, then more coverage is always progress, and the absence of results means the threshold has not yet been reached rather than that the model was wrong. Like the knowledge panel claims examined elsewhere in this series, the arrangement has no failure condition.

There is a further attraction in its unfalsifiability. Because no score is published, no one can demonstrate that a campaign failed to raise it. Where a claim cannot be checked, the claim survives every outcome.

None of this means people using the phrase are being dishonest. Most use it loosely, to mean covering a subject properly, which is good advice. The problem is the drift from a description of good practice to a quantity with a price attached.

It survives partly because the honest version and the marketed version share a vocabulary. An editor saying a publication has authority on a subject means something real and qualitative. A vendor using the same words to describe a score being moved by a purchase means something else entirely, and the sentence sounds identical.

Can it be measured?

Not as a quantity inside Google, because no such published quantity exists. Third-party tools offering topical authority scores are in the same position as those offering domain authority: producing a model's output from externally observable signals, not reading a value from the ranking system.

What can be measured are consequences rather than the supposed cause. How many queries in a subject area does a site rank for, and has that set grown? Do its pages hold position over time? Do other publications cite it when covering the subject? Do readers arrive for one article and return for others?

These are real measurements, available without any vendor's proprietary score, and they answer the question a buyer actually has — whether the investment is producing coverage of a subject that people and systems find useful.

The distinction matters practically. A campaign reported against a vendor's authority score is reported against a number nobody outside that vendor can verify. A campaign reported against query coverage and retained rankings is reported against outcomes anyone can check.

There is one measurement worth adding, because it is the one buyers most often skip: whether the coverage is read. Time on page, returning visitors and onward navigation say whether a body of work is doing anything for the people it was written for, and a subject programme that no one reads is not building authority of any kind, measurable or otherwise.

What this means for anyone buying placement

Treat topical authority as a description of a strategy, not as a deliverable. Covering a subject substantively is sound practice and will tend to help, for ordinary reasons that require no hidden score.

Refuse the metric version. If a proposal promises to raise a topical authority score, ask which system that score belongs to and where its operator documents it. The answer will be a third-party tool, which is fine as a tool and useless as a target.

Ask for outcome measures instead: the queries a body of work ranks for, whether those positions persist, and whether anyone independent cites the material. Those can be checked by you rather than reported to you.

Media Morpheus sells placement and search visibility work. Publishing substantively about a subject is a reasonable strategy and part of what we do. There is no topical authority score to buy, no threshold we can move you across, and any seller quoting a figure for your subject authority is quoting a number that exists only inside the tool that produced it.

Methodology and limitations

This report relies on Google's published guide to its Search ranking systems, examined for the systems it names and the language it uses about page-level and site-wide signals. Four limits apply. First, the absence of a topical authority system from Google's documentation is evidence about what Google has published, not proof about the internals of a system Google does not fully describe. Second, no third-party tool's scoring methodology is evaluated here, and nothing in this report asserts that such scores are useless for the comparative purposes they were built for. Third, no correlation data between subject-focused publishing and ranking outcomes is cited, because published studies on that question are produced largely by firms selling content programmes. Fourth, platform documentation changes without notice and this describes the position at the date of publication.

Conclusion

Topical authority is a good description of a sound practice and a poor description of a mechanism. Google's own guide names seventeen ranking systems, none of which is a topical authority system, and the site-level language it does use points the other way — ranking decided page by page, with site-wide signals contributing to understanding and guaranteeing nothing. Covering a subject thoroughly remains worth doing, because it produces work that people cite, link and return to. What cannot be done is measuring the resulting authority as a number, crossing a threshold, or buying a quantity of it, and a campaign priced against a score no one can verify is priced against nothing.

Frequently Asked Questions

Sources

1. official Google Search Central. A guide to Google Search ranking systems, including the list of current and retired systems and the description of page-level and site-wide signals.