Skip to content
Learn

Brand signals and entity consistency

In this lesson

Section titled “In this lesson”
  • Explain why search engines and AI systems compare what many sites say about one brand or person.
  • Find and fix mismatches in names, bios and details across your site, profiles and directories.
  • Tell which off-site sources confirm identity and which show reputation.
  • Decide what to do with an unlinked mention of your brand.
  • Take part in forums and communities without breaking platform rules or Google’s spam policies.

Day 06 was about links. This lesson is about signals that do not need a link. Other sites describe you in directories, profiles, news articles and forum threads. Search engines and AI systems read those descriptions. They use them to confirm who you are and what people think of you.

Why brand signals matter

Section titled “Why brand signals matter”

On day 05b, you learned that Google stores entities (people, places, organizations and things) in its Knowledge Graph. Google’s knowledge panel help page says that panel information “comes from various sources across the web”. Some data also comes from partners, and some from the entities themselves.

So the description of your brand is not only your own. It is the sum of many descriptions on many sites. This gives two jobs:

  1. Identity. All the descriptions must clearly point to the same entity. This is entity consistency.
  2. Reputation. Independent sources must say good things about that entity. This comes from mentions, reviews and community talk.

Google’s Search Quality Rater Guidelines (September 2025) show why the second job matters. Raters are the people who check the quality of Google’s results. The guidelines tell them: “When there is disagreement between what the website or content creators say about themselves and what reputable independent sources say, trust the independent sources.” Rater scores do not change rankings directly. But they show what Google’s systems try to reward.

The roadmap.sh items add a claim: AI systems also use these signals when they choose which brands to name. Correlation data supports this (see “Unlinked brand mentions” below), but no AI company has confirmed how its system weighs them. Day 07 covers AI search in detail.

Entity consistency

Section titled “Entity consistency”

Entity consistency means that your name and main facts are the same everywhere they appear: your site, your profiles, directories and other people’s articles. When the details match, a system can join the scattered mentions into one entity. When they do not match, the system may see two or three different companies, or it may not be sure.

Google’s documentation gives several direct rules for your own site:

  • Site name. Google’s site names page says: “Use your site name consistently across your home page.” The WebSite structured data “is most important”, but Google also reads the <title>, headings and og:site_name.
  • Organization markup. Google asks you to use “the same name and alternateName” in Organization markup as in your site name. It says that this markup helps Google “disambiguate your organization in search results”.
  • Business Profile. For local businesses, Google says: “Represent your business as it’s consistently represented and recognized in the real world across signage, stationery, and other branding.” Local details such as name, address and phone number come on day 09.

What must match? Start with the facts that identify you: the name, the logo, the home page URL, the founder or main people, and the location. Then check the description. A one-line description that is the same on every profile helps a system and a reader.

Common causes of mismatch are a rebrand, a move to a new office, a new domain, or many people who each wrote their own version of the bio.

Use the website name "Acme Tools" on:

Toy model with example data. The website says 'Acme Tools'. Tick each place to fix its name. Notice how each fixed place joins the brand with a solid line, and the readout lists the places that still disagree.

The toy model counts only names. Real systems compare many attributes at once, so one small difference is not fatal. But each mismatch makes the join less certain. Fix the places that you control first: your site, your markup and your profiles.

Author and organization consistency

Section titled “Author and organization consistency”

Author and organization consistency applies the same idea to people. An author writes on your blog, on guest posts and on a professional profile. The name, job title and bio must match in all of these places.

Google’s helpful content guide asks: “Do pages carry a byline, where one might be expected? Do bylines lead to further information about the author?” It also warns against “fabricating creator profiles”, such as “AI-generated headshots, made-up names, or false credentials”. Google calls that deception and “a signal of a low-quality page”.

Google’s Article markup rules show how to connect an author to one entity:

  • Put only the name in author.name. Do not add the job title, the publisher or words like “posted by”. Use jobTitle and publisher for those.
  • Add url or sameAs that points to “a web page that uniquely identifies the author”, such as a bio page or a social profile. Google says it “can understand both sameAs and url when disambiguating authors”.
  • Use the Person type for people and the Organization type for organizations.

A good pattern is one author page on your site for each writer. Use the same name and photo on that page, in the byline, in the markup and on the writer’s profiles. Then link the author page to those profiles.

Directories and databases

Section titled “Directories and databases”

A directory is a list of businesses or people, often for one industry or one place. A database here means a structured, public record about entities, such as Wikidata. These listings matter because they are outside your site. They confirm your details independently, beyond what you say about yourself.

Not all listings are equal:

Kind Example Value
Open knowledge base Wikidata High for identity. Day 05b showed that knowledge graphs use it.
Trusted industry or official register A professional body, a chamber of commerce Confirms that you exist and are in good standing
Business data platform Crunchbase, a review site Confirms facts such as founders and location
Low-quality link directory A site that lists anyone who pays None. Google’s spam policy names “low-quality directory or bookmark site links” as link spam

Wikidata has rules. Its notability policy accepts an item only if it links to a Wikimedia page, describes “a clearly identifiable conceptual or material entity” with “serious and publicly available references”, or fills a structural need. So you cannot add your brand only because you want a listing. You must have public references first.

For each directory, keep the details the same as on your site. A listing with an old name or address is a source of mismatch, not a confirmation.

Social profiles

Section titled “Social profiles”

Social profiles are your accounts on platforms such as LinkedIn, X, YouTube and Facebook. They do two jobs for identity.

First, they connect to your site. List them in the sameAs property of your Organization or Person markup. Google defines sameAs as “the URL of a page on another website with additional information about your organization”. Link back from each profile to your site, so the connection goes both ways.

Second, they help you prove who you are. To claim a knowledge panel, Google asks you to “sign in to one of the official sites or profiles listed”, for example YouTube, Search Console, Twitter (now X) or Facebook. After you claim it, you can suggest changes to the facts.

Social profiles have a limit. The rater guidelines say that a company’s “official social media pages that it closely maintains” are “not … independent sources of reputation information”. So a profile confirms who you are. It does not prove that people trust you.

Unlinked brand mentions

Section titled “Unlinked brand mentions”

An unlinked brand mention is a place where another site names your brand, but does not link to it. These mentions are common. A journalist names a product, or a blogger lists tools that they use.

Do they help in search? The answer has two parts:

  • Not as links. In 2022, Google’s John Mueller answered this question in a Search Central office-hours video: “we don’t treat these as links”. So a mention passes no PageRank.
  • As reputation evidence. The rater guidelines tell raters to search for a brand on other sites and to read “news articles, Wikipedia articles, blog posts, magazine articles, forum discussions”. A mention can show reputation even without a link.

For AI answers, a large study shows a strong correlation. In May 2025, Ahrefs studied about 75,000 brands. Branded web mentions had the strongest correlation with mentions in Google’s AI Overviews (0.664). Backlinks had a much weaker one (0.218). This is correlation, not proof of cause. Big brands get more mentions and more AI visibility for the same reason: they are big.

What to do with an unlinked mention:

  1. Find them. Set up Google Alerts for your brand name. Google says that alerts can send you “mentions of your name”. Paid tools such as Ahrefs Content Explorer also find them.
  2. Check the facts. If the mention has wrong details, ask for a correction first.
  3. Ask for a link when it helps the reader. Ahrefs’ guide says that the writer “is familiar with your business”, so you are “already halfway towards earning a link”. Be polite and short. Accept a “no”.

Do not pay for the link. A paid link without rel="sponsored" is link spam, as day 06 showed.

Community presence and niche forums

Section titled “Community presence and niche forums”

Community presence is your activity and reputation in online communities, such as Reddit, Quora and forums. A niche forum is a smaller community about one industry, hobby or interest. Niche forums often have the most detailed answers in their field, for example a forum for one camera brand or one programming tool.

These communities are now more visible in search:

  • In May 2023, Google added a “Perspectives” filter. It highlights posts from discussion boards, Q&A sites and social media. Google’s docs now name a “Discussions and forums” feature.
  • In November 2023, Google added DiscussionForumPosting markup for forum sites. Google says that it helps “identify online discussions across the web” for those features. If you run a forum, add it.
  • In February 2024, Google announced an expanded partnership with Reddit. Google got access to Reddit’s Data API, to “display, train on, and otherwise use” Reddit content.

AI tools also cite these communities. A Semrush study of 230,000 prompts (July to October 2025) found that Reddit and LinkedIn were in the top five most-cited domains on ChatGPT search, Google AI Mode and Perplexity. The same study showed big swings: ChatGPT cited Reddit much less after mid-September 2025. So community citations are useful, but they are not stable.

How to take part without harm:

  • Follow the rules. Reddit’s rule 2 says: “Participate authentically in communities where you have a personal interest, and do not spam”. Each community also has its own rules.
  • Say who you are. The US Federal Trade Commission (FTC) says that an employee who mentions the company’s products on social media “should disclose your relationship to the company”. “Listing your employer on your profile page isn’t enough.”
  • Do not drop links. Google’s spam policy names “forum comments with optimized links in the post or signature” as link spam.
  • Answer first. Help with the question. Name your product only when it really fits, and name other good options too.
Your action

Pick a situation, then an action. Notice that the best action changes with the situation, and that paying for a link is never the best choice.

What matters most

Section titled “What matters most”

Not all of these items need the same effort. Use this order:

Priority Item Why
1 Consistency on your own site, markup and profiles You control it, it is fast to fix, and every other signal builds on it
2 Earned mentions in independent sources Raters and, by correlation, AI answers rely on what others say
3 Honest community presence High reach and many AI citations, but it takes time and must follow each community’s rules
4 Directories and databases Useful for identity, but only the trusted ones. Paid link directories are a risk

The main idea: you cannot fully control what the web says about you. You can make the facts easy to match, and you can give people real reasons to talk about you.

Key terms

Section titled “Key terms”
  • Brand signal: anything outside your site that tells a system who you are or what people think of you.
  • Entity consistency: using the same name and main facts for one entity on every site where it appears.
  • Author and organization consistency: entity consistency for authors and companies, such as names, job titles and bios.
  • Directory: a list of businesses or people, often for one industry or one place.
  • Database: here, a structured public record about entities, such as Wikidata.
  • Social profile: an account on a platform such as LinkedIn, X or YouTube.
  • sameAs: a schema.org property that lists other pages about the same entity.
  • Unlinked brand mention: a place where another site names your brand without a link to your site.
  • Reputation research: checking what independent sources say about a site or creator.
  • Community presence: your activity and reputation in online communities, such as Reddit and forums.
  • Niche forum: a small online community about one industry, hobby or interest.

Exercise

Section titled “Exercise”

Exercise · about 15 min

Audit the brand signals of one brand

Pick a brand: your own, a client, or a well-known one such as Moz (moz.com).

Part 1: what the brand says about itself.

  1. Open the home page. View the page source (Ctrl+U, or Cmd+Option+U on a Mac).
  2. Search for sameAs. Write down each profile URL in the list.
  3. Open two of those profiles. Write down the name, the description and the link back to the site. Do they match the home page?

Part 2: what others say.

  1. Search Google for moz -site:moz.com. This is a search from the rater guidelines. Use your brand and domain instead.
  2. For the first ten results, write down: who controls the page (the brand or someone else), and whether it links to the brand.
  3. Search moz site:reddit.com. Read one thread. Write one sentence about what people say.
  4. Decide: which one fix or action would help this brand most?
Show a possible solution

Checked on 14 October 2026: the Moz home page has Organization markup with a sameAs list of five profiles: Twitter (twitter.com/Moz), Facebook, LinkedIn, YouTube and Instagram. The markup also lists two addresses, Seattle and Vancouver. The <title> starts with “Moz”, the same name as the markup.

One small finding: the sameAs list still uses a twitter.com URL, the old address of X. The URL still works, but x.com is the current address.

In part 2, expect a mix. Moz’s own profiles (LinkedIn, YouTube, X) are controlled by the brand, so they confirm identity but do not show reputation. Wikipedia, news articles, reviews and forum threads are independent. Many of them name Moz without a link: these are unlinked mentions. A good action for a smaller brand is often to fix the mismatches in part 1 first, then ask for links from the friendliest unlinked mentions in part 2.

Check yourself

Section titled “Check yourself”
A company rebranded from 'Brightline' to 'Brightway' a year ago. Its site uses the new name, but LinkedIn, two directories and many old articles use the old name. What is the best first step?
A tech blog writes 'We tested Brightway and it was fast' with no link. Based on Google's statements, what does this mention do?
Which source is the strongest evidence of a brand's reputation for a rater?
A marketing manager wants to answer Reddit questions about tools in their field. Which plan follows Reddit's rules, the FTC guidance and Google's spam policies?
An Ahrefs study found a 0.664 correlation between branded web mentions and mentions in AI Overviews. What is the correct conclusion?

Sources

Section titled “Sources”

Next

Section titled “Next”

Day 07 covers AI search and GEO: how AI Overviews, AI Mode, ChatGPT and Perplexity find, choose and cite sources, and how to manage AI crawlers.