Named Author Markup: The AI Citation Signal Most Service Businesses Skip

Most service business websites are published by nobody. The about page says "our team." The services page says "we offer." The blog, if there is one, lists no author. From an AI perspective, the content exists but it belongs to no one.

Two independent studies from 2026 suggest that's a citation problem -- and it's fixable with a specific, underused markup change.

What the research found

In our October 8, 2026 update to `knowledge/content-format-citation-signals.md` (Scout session 162), we documented the Visionary Marketing AEO Benchmark Study: a cross-engine analysis of 12,400 AI-search-eligible queries across ChatGPT, Perplexity, Google AI Overviews, and Gemini, paired with a 100,000-page ranking factor crawl.

One finding: pages with named authors that have `sameAs` links to professional profiles -- LinkedIn, association directories, biography pages -- earned citations at 39% higher rates than equivalent pages with anonymous or absent authorship. The study was published April 2026 and covered a mixed-quality query set, which is why the lift figure is more conservative than what we see in other datasets.

The corroborating data comes from a significantly larger source. Conductor's 2026 AEO/GEO Benchmarks Report, which covers 13,770 enterprise domains and 3.3 billion sessions, found author entity markup and person schema improved citation likelihood by 67%. We documented this in `knowledge/schema-markup-effects.md` (Scout session 44, 2026-06-11). Conductor's dataset skews toward higher-authority enterprise domains; Visionary Marketing's is broader and more mixed. That gap explains the 39% vs. 67% difference. Both point in the same direction.

Conductor's framing of the mechanism is worth noting: person entities help AI systems connect expertise, authorship, and topical trust. A service page attributed to a licensed electrician named Maria Chen, linked to her Electrical Contractors Association profile and LinkedIn, sends different signals than an identical page with no attribution. Not because the name is magic -- because the linked entity is verifiable.

Why attribution matters to AI systems

AI models are trained on text where attribution is the norm. Journalism cites sources. Academic papers list authors. Wikipedia attributes claims to references. Anonymous content -- "our team of experienced contractors" -- reads as lower-credibility at the model level because it doesn't match the authorship patterns in the training data that AI systems treat as authoritative.

The SIGIR 2026 study (arXiv:2605.25517), accepted at ACM SIGIR 2026, ran 252,000 controlled citation trials across multiple AI platforms and built an 18-factor taxonomy of what actually moves citation probability. Their primary finding: topical relevance and ranking position dominate. Structural signals -- schema, author markup, heading format -- are secondary amplifiers, not primary drivers. We documented this in `knowledge/content-format-citation-signals.md` (Scout session 71, 2026-07-08).

The ordering matters for how to think about author markup. It won't rescue a page that isn't relevant to what buyers actually ask, and it won't overcome weak organic rankings. But for a page that has cleared those bars -- it's ranking, it's topically matched to real queries -- the 39-67% author entity lift represents a meaningful secondary gain that most businesses are leaving on the table.

What "named author with sameAs links" actually means

A name in a page footer doesn't accomplish much. What the research measures is a connected entity: a named person whose identity is verifiable across multiple external profiles.

The markup is a `Person` schema block in JSON-LD. For a single-location service business, it looks like this:

```json { "@context": "https://schema.org", "@type": "Person", "name": "Maria Chen", "jobTitle": "Licensed Master Electrician", "sameAs": [ "https://www.linkedin.com/in/mariachen-electric", "https://eca.ca/member-directory/maria-chen", "https://contractors.ontario.ca/license/XXXXX" ], "worksFor": { "@type": "Electrician", "name": "Chen Electrical Services" } } ```

The `sameAs` URLs are doing the work. They give AI systems traversal paths to the same entity record across multiple external data sources. Without the links, "Maria Chen" is a string. With links to her ECA directory entry, her LinkedIn profile, and her provincial license listing, she's an entity with external verification points. That's what allows a model to reason about expertise and credibility, not just surface a name.

The `sameAs` URLs need to point to real, publicly accessible profiles. A link to a locked profile or a broken URL doesn't create a traversal path. A professional association directory entry with her name, credentials, and location does.

Where to apply it

**Service pages.** If your business is built around licensed or credentialed professionals -- plumbers, electricians, dentists, lawyers, accountants, contractors -- the person who holds the credential is the natural author of the service page. Their name and credential, linked to their professional profile, is the attribution.

**Blog and FAQ content.** Any post that answers a question buyers ask AI engines is stronger with a named author. "How to know if your AC compressor needs replacement" attributed to a certified HVAC technician reads differently to an AI retrieval system than the same post with no author.

**About and staff pages.** These are the most obvious and the most underutilized. A staff page that lists names without markup is a missed entity-establishment opportunity. Each named professional should have a `Person` entity with at least one `sameAs` link to an external verifiable profile.

For a single-location service business, the practical priority: start with the service page that drives your most valuable queries, add the `Person` entity for the most credentialed person at the business, and check whether citations shift over 60 days. Don't run it as a blanket site-wide change before you have a baseline.

What this doesn't fix

Author markup won't compensate for a page that isn't topically matched to what buyers actually ask. A dentist's whitening page that buries the word "whitening" below the fold, uses professional category language instead of buyer language, and isn't ranking organically -- a `Person` schema block changes nothing meaningful. The SIGIR hierarchy is real: relevance first, position second, structural signals third.

It also won't fix fragmented entity data. If your `Organization` schema uses a different business name than your Google Business Profile, and your GBP uses a different address format than your Yelp listing, an author entity sits on top of a fragmented foundation and can't resolve the underlying problem.

The right sequence: entity consistency first (name, address, phone consistent across GBP, Yelp, and schema), then content structure, then author markup. Author markup is a Phase 2 amplifier -- meaningful once the basics are clean.

Checking your current baseline

Before changing anything, it helps to know whether your pages are currently being cited at all, and on which platforms the gaps are. Signal Check at sourcepull.ca runs a cross-platform visibility test across ChatGPT, Perplexity, Gemini, and Claude in about 60 seconds. The output shows per-platform citation rates and a prioritized fix plan -- so you can tell whether you're dealing with an entity problem, a content gap, or both. Free, no account required.