When Google can’t tell it’s all you
Structure and markup refreshed for current answer engines; the original analysis is preserved.
Google does not just index pages; it builds a model of the entities behind them — your company, the people in it, the things you make — by gathering signals from across the web. When those signals are consistent and explicitly connected, it forms one confident picture of who you are. When they are scattered, contradictory, or unlinked, it can fragment your identity or confuse you with someone else. Entity markup and consistent references are how you tell the engine that all of it is you.
the short answer
Google models the entities behind your pages — your company, people, products — by gathering signals from across the web. Scattered or contradictory signals fragment you or confuse you with a namesake. The sameAs property and consistent references let you state, rather than make Google infer, that your profiles and listings are you — so it forms one confident entity instead of guessing.
key takeaways
- Google does not just index pages — it builds a model of the entities behind them (your company, your people, your products) by gathering signals from across the web into one picture.
- That picture can fail two ways: fragmentation (scattered, unlinked signals never merge into one confident entity) or confusion (a similarly named other gets blended with you). Both come from leaving Google to guess.
- The sameAs property lets you state, in your Organization or Person markup, that your official profiles and listings are you — turning an inference Google has to make into a fact you have given it.
- Consistency is what lets signals merge: the same name, address, and description across your site, profiles, and directories. Inconsistency is what forces the engine to guess.
- Mark up your entity, claim and align the profiles you list in sameAs, and treat your own site as the canonical source of truth. Don’t make Google infer what you can simply state.
scattered signals vs. one stated identity
The signals are the same on both sides. What changes the outcome is whether you leave Google to connect them or connect them yourself — explicitly, in markup and in consistent references.
The idea, in four parts
How Google models entities, not just pages; how the picture breaks; how markup connects you to yourself; and what to do about it. Open each part.
01 Google models entities, not just pages
It is easy to picture Google as an index of pages and the words on them, but that is only half of what it keeps. Alongside the pages, Google maintains a model of the real-world things those pages are about — your company, the people who run it, the products you make — and it has done so for years through its Knowledge Graph, a vast store of entities and the facts and relationships attached to each: who founded an organization, where it is based, what it produces, which other entities it connects to. So when someone searches, Google is not only matching a query to a page. It is also asking which entity this concerns and what it already knows about that entity, and the answer shapes what it shows. A search for your brand can pull up a Knowledge Panel drawn straight from this model, and your pages can rank partly because Google recognizes and trusts the entity standing behind them. The part that matters most for your work is where this model comes from. Google does not receive a tidy file describing your organization; it assembles the picture itself, gathering signals from across the web — your own site, your profiles on other platforms, mentions and citations, the structured data you publish — and reconciling them into a single understanding. That assembled picture can be accurate and confident or partial and muddled, and which one you get depends almost entirely on how clear and consistent the signals you emit are. Your entity is not just discovered; it is, in large part, constructed from what you make available.
02 How the picture breaks
There are two main ways the engine ends up with the wrong picture of you, and both are common. The first is fragmentation. The signals about you exist, but they are scattered and never explicitly tied together: your site uses one version of your name, your social profiles use a slightly different one, your address appears in three formats across directories, and nothing anywhere states that the company on your About page is the same company as the LinkedIn profile and the trade listing. A person glancing at all of these would assume they describe one organization. Google does not assume — it infers from evidence, and scattered, inconsistent evidence yields a weak, fragmented model, or several partial ones that never merge into the single, confident entity that earns a Knowledge Panel and lends weight to your pages. The second failure is confusion. If another company or person carries your name, or one close to it, and the signals that should tell you apart are thin, Google can blend the two identities or hang one’s facts on the other — your achievements credited elsewhere, or someone else’s baggage attached to you. Neither outcome is unusual, and both trace back to the same root: the engine was left to guess because the connections it needed were never made explicit. The fix is implied by the cause. Wherever Google is guessing about your identity, you have the option to stop making it guess and simply state the answer instead.
03 How markup connects you to yourself
The most direct tool for stating your identity is structured data, and the most pointed part of it for this purpose is the sameAs property. In your Organization or Person markup, sameAs lets you list the authoritative URLs that represent you elsewhere — your official social profiles, your entry in industry databases, your Wikipedia or Wikidata page if one exists — and in doing so declare, plainly, that those pages are the same entity as the one your site describes. Each sameAs link converts an inference into a statement. Rather than leaving Google to deduce that your About page, your LinkedIn, and your directory listing all refer to one organization, you tell it so, and it can merge those scattered signals into a single entity with far greater confidence. Confidence is the currency the whole model runs on: a well-connected, confidently understood entity is the one that earns an accurate Knowledge Panel, gets disambiguated correctly from others who share its name, and has its authority recognized when its pages compete. It is worth being honest about what sameAs is not. It is not a ranking lever, and it will not manufacture authority you have not earned; what it does is ensure that the identity and authority you genuinely have are attributed to the right entity — you — instead of being scattered, lost, or handed to a namesake. Supported by consistent naming and a stable identifier for your entity across your markup, sameAs turns a pile of loosely related signals into one recognizable you. Using markup to make identity explicit, so that nothing true about you goes unconnected, is the discipline the AC Group has worked by for {years} years.
04 What to do about it
Approach your entity as something you construct on purpose rather than something you hope the engine reconstructs correctly on its own. Begin with markup: describe your organization, and the key people in it, using Organization and Person structured data on your own site, give each a stable identifier so you can refer to the same entity consistently across your pages, and include a sameAs list pointing to the authoritative profiles and listings that represent you elsewhere. Next, make your references consistent wherever they appear — the same legal name, the same address formatting, the same core description — across your site, your social profiles, and every directory or database that mentions you, because inconsistency is precisely what forces Google to guess and consistency is what lets it merge. Claim and align the profiles you intend to list in sameAs, so that when Google follows each link the destination confirms the same identity rather than introducing a new contradiction. And treat your own website as the canonical source of truth about your entity — the authoritative place the definitive facts live and that everything else points back toward. None of this is about gaming the engine; it is about removing the ambiguity that makes it form a poor or split picture of you, so that the accurate, connected, confident version is the one it ends up holding. Building a clear, well-connected entity the engine can recognize as a single, coherent you is exactly the entity and schema work the AC Group has done for {years} years.
Why this is an identity point, not a markup chore
It is tempting to treat entity markup as a technical checkbox — add some structured data, tick the box, move on. But the markup is only the mechanism; the real subject is your identity, and whether the engine holds an accurate, connected version of it or a vague, split one. That distinction has consequences well beyond a Knowledge Panel. The entity Google forms of you is the thing it carries from one query to the next, the lens through which it reads your pages and weighs whether to trust them. If that entity is fragmented or confused with a namesake, every page you publish inherits the ambiguity; if it is clear and confident, your pages inherit that standing instead.
So the work is not really about the tags. It is about refusing to leave your identity to chance — making sure that everything true about you is connected, attributed, and unambiguous, so the engine has no room to assemble a worse version than the real one. That is a discipline of clarity more than of code: state plainly what you are and how your scattered presences relate, and let nothing important stay an inference. Making identity explicit and well-connected, so the engine recognizes a single coherent you, is the entity and schema work the AC Group has done for 27 years.
What to do with this
Build your entity on purpose. Mark up your organization and your key people with Organization and Person structured data, give each a stable identifier so the same entity is referenced consistently across your site, and include a sameAs list pointing to the authoritative profiles and listings that represent you elsewhere. Then make every reference to you consistent — the same name, address formatting, and core description — across your site, your profiles, and every directory that lists you, because inconsistency is what forces Google to guess and consistency is what lets it merge the signals into one.
Claim and align the profiles you list in sameAs, so each link confirms your identity rather than contradicting it, and treat your own site as the canonical source of truth that everything else points back to. None of this games the engine; it removes the ambiguity that makes it form a poor or split picture, so the accurate, connected version is the one it keeps. Building a clear, well-connected entity the engine can recognize as a single, coherent you is the entity and schema work the AC Group has done for 27 years.
Entities and sameAs, plainly: quick answers
What does it mean that Google understands entities, not just pages?
It means Google does not only keep a list of your web pages and the words on them; it also tries to identify the real-world things those pages are about — your company, the people who work there, the products you sell — and to hold a single, connected understanding of each one. These real-world things are entities, and Google has maintained a Knowledge Graph of them for years, linking each entity to the facts and relationships it has gathered: who founded a company, where it is based, what it makes, which other entities it is connected to. The practical upshot is that Google is not just asking which page matches this query. It is also asking which entity is this about, and what do I know about that entity. A search for your brand can surface a Knowledge Panel assembled from this model, and your pages can rank in part because Google understands the entity behind them and trusts it. The important thing to grasp is where the model comes from: Google builds it by gathering signals from across the web — your own site, your profiles on other platforms, mentions and citations, structured data — and reconciling them into one picture. That picture can be accurate and confident, or partial and confused, depending entirely on how clear and consistent the signals you put out are. Shaping those signals so the engine forms the right picture of you is the core of entity work.
How can Google get my entity wrong?
Mostly in two ways: by fragmenting you, or by confusing you with someone else. Fragmentation happens when the signals about you are scattered and never explicitly connected. Your website calls you one name, your social profiles use a slightly different one, your address is formatted three ways across directories, and nothing in your markup tells Google that the company on your About page is the same company as the LinkedIn profile and the industry listing. A human reader would assume they are the same; Google does not assume, it infers from evidence, and weak or inconsistent evidence produces a weak, fragmented model — or several partial ones that never merge into the confident single entity that earns a Knowledge Panel and lends authority to your pages. Confusion is the other failure: if another organization or person shares your name, or a similar one, and the distinguishing signals are thin, Google can blend the two or attribute one’s facts to the other. Neither failure is exotic; both are common, and both come from the same root cause — the engine was left to guess because the connections it needed were never made explicit. The good news is that the same root cause points at the fix: stop making Google infer what you can simply state. Stating it clearly, in markup and in consistent references, is the discipline the AC Group has worked by for 27 years.
What is sameAs and why does it matter?
The sameAs property is a piece of structured data that lets you tell Google, explicitly, that a set of other pages refer to the same entity as your own — a direct way of saying these profiles are me. In your Organization or Person markup, you list the authoritative URLs that represent you elsewhere: your official social profiles, your entry in industry databases, your Wikipedia or Wikidata page if you have one. Each sameAs link is a stated connection rather than an inferred one, and that is exactly what resolves the fragmentation problem. Instead of hoping Google works out that your About page, your LinkedIn, and your directory listing are all the same organization, you tell it so directly, and it can merge those signals into one entity with much higher confidence. This matters because confidence is what the whole entity model runs on. A confident, well-connected entity is far more likely to earn an accurate Knowledge Panel, to be disambiguated correctly from similarly named others, and to have its authority recognized when its pages compete to rank. sameAs is not a magic ranking switch and it will not invent authority you have not earned; what it does is make sure the authority and identity you do have are attributed to the right entity — you — rather than lost or misassigned. Using markup to make identity explicit, so nothing true about you goes unconnected, is the discipline the AC Group has worked by for 27 years.
What should I actually do?
Treat your entity as something you build deliberately, not something you hope Google guesses right. Start by marking up your organization — and the key people in it — with Organization and Person structured data on your site, giving each a stable identifier and including a sameAs list that points to the authoritative profiles and listings that represent you elsewhere. Then make your references consistent everywhere they appear: use the same legal name, the same formatting for your address and contact details, the same core description, across your own site, your social profiles, and every directory or database that lists you. Inconsistency is what forces Google to guess; consistency is what lets it merge. Claim and align the profiles that matter — the ones you will list in sameAs — so that when Google follows the link, the destination confirms the same identity rather than muddying it. And treat your own site as the canonical source of truth about your entity: the place where the definitive facts live, that everything else points back to. None of this is about tricking the engine; it is about removing the ambiguity that makes it form a poor picture of you, so the accurate, connected, confident version is the one it holds. Building a clear, well-connected entity that the engine can recognize as a single you is exactly the kind of entity and schema work the AC Group has done for 27 years.
A note on sources and timing
This is written in June 2020. The behaviour described — that Google maintains a Knowledge Graph of entities, builds its understanding of each entity by reconciling signals from across the web, and uses the sameAs property in Organization and Person structured data to connect a site to the profiles and listings that represent the same entity — follows Google’s long-standing documentation on structured data and entities. The reading offered here — that the practical risks are fragmentation and confusion, and that the remedy is explicit connection plus consistent references — is our interpretation, grounded in that documented behaviour. The durable point holds regardless of how the tooling evolves: if you leave your identity to be inferred, the engine may infer it wrong — so state it. Making identity explicit and well-connected is the discipline the AC Group has worked by for 27 years.