The engine might think you’re someone else
Structure and markup refreshed for current answer engines; the original analysis is preserved.
If another organisation shares your name, the Knowledge Graph can confuse the two of you — blending your signals, mixing your knowledge panel, and crediting your hard-won authority to your namesake. Disambiguation is how you tell the engine which entity you are, so the authority you earn is attributed to you.
the short answer
The Knowledge Graph thinks in entities, and has to decide which entity a name means. If another organisation shares your name, it can confuse the two of you — blending signals, mixing your knowledge panel, crediting your authority to your namesake. Disambiguation is how you tell the engine which entity you are: sameAs links to authoritative profiles, plus the unique attributes only you have. It is not a ranking trick — it makes you legible, so the authority you earned is filed under your name.
key takeaways
- The Knowledge Graph stores the world as entities, not just words, and has to decide which entity a name refers to.
- If another organisation shares your name, the engine can confuse the two of you — blending signals, mixing your knowledge panel, crediting your authority to your namesake.
- Disambiguation is how you tell the engine which entity you are: sameAs links to authoritative profiles, plus the unique attributes only you have.
- It matters because authority attaches to an entity. With the wrong entity, the citations and mentions you earn are filed under someone else.
- It is not a ranking trick. It makes you legible, so the authority you genuinely earned is seen as yours when the system re-evaluates.
confused vs resolved
On the left, two organisations collapse into one node the engine cannot trust; on the right, a handful of unambiguous signals pull them apart. The difference is not how good your content is — it is whether the engine can tell which of you it is looking at.
Why this is an identity problem, not a content problem
It is easy to misread the symptoms of entity confusion as a content or authority shortfall and respond by writing more, building more links, trying harder to deserve the panel you are not getting. But if the underlying issue is that the engine cannot tell you from your namesake, none of that effort resolves it — some of it may even flow to the other entity. The tell is in the shape of the problem: a knowledge panel that mixes facts, mentions that do not seem to count, a brand the engine treats as hazy despite real activity. Those are not signs that you have done too little; they are signs that what you have done is being filed under an identity the engine has not pinned down. The fix is upstream of content entirely — it is making the identity unambiguous first, so the rest can attach to it.
This is also why disambiguation is not a ranking trick and should not be sold as one. Telling the engine which entity you are does not, by itself, make you more authoritative or push you up the results; it simply ensures that the authority you genuinely have is credited to you rather than leaking to a namesake. With the core update rolling out now re-evaluating quality across the web, that legibility matters more than usual, because a reassessment can only reward the right entity if it knows which entity is right. Resolving your identity is the unglamorous groundwork that lets everything else — your content, your authority, your reassessment — land where it belongs, which is the kind of quiet, structural correctness — making sure the foundations say who you are before you build on them — that the AC Group has worked at for 27 years.
The argument, in three parts
The engine thinks in entities and names are ambiguous; confusion with a namesake costs you concretely; and you resolve it by giving unambiguous, corroborated signals. Open each part.
01 The engine thinks in entities, and names are ambiguous
The Knowledge Graph does not store the web as a pile of words; it stores it as entities — distinct organisations, people, places, and things — and the relationships between them. That is what lets a search for "Apple" resolve to the company rather than the fruit or the record label: the engine is disambiguating a shared name down to one entity using the surrounding context and the signals it trusts. The same machinery runs on your brand. When your name is unique, resolution is easy and invisible. When you share it with another organisation, the engine faces a genuine decision about which of you a given page, mention, or query refers to — and nothing guarantees it decides in your favour rather than handing the recognition to whichever organisation it happens to know better. The first thing to understand is that your name is not your identity to the engine; the entity behind it is, and that entity has to be made clear deliberately, because the engine will not always guess it for you.
02 What confusion actually costs you
When the engine cannot cleanly separate you from a namesake, the costs are concrete rather than abstract. Your knowledge panel — the box the engine builds for your brand, effectively a second homepage you do not fully control — can show their founding date, their logo, their facts mixed with yours. Signals split: a mention that should strengthen your standing is attributed to the other entity, so the reputation you are building partly accrues to someone else, who did nothing whatever to earn it except happen to share your name. In the worst case the engine simply hedges, showing a thin or absent panel because it is not confident which entity is real. None of this is a content problem you can write your way out of; it is an identity problem, and it persists until the identity is resolved, quietly taxing everything else you do until you address the identity directly.
03 How to resolve your identity
Resolving it is a matter of giving the engine enough unambiguous, corroborated signal to hold you as one entity. The anchor is the sameAs property in your Organization markup: link to the authoritative profiles that are unmistakably you — verified social accounts, a Wikidata entry, official directory listings — since each is a corroboration that this site and that profile are the same thing. Reinforce it with the attributes only you possess: founding date, location, your specific field, a clear About page, and a name, address, and phone that stay consistent everywhere. Earn third-party mentions that tie your exact name to your site. None of this fabricates authority; it makes you a single, legible node, so that when the engine re-evaluates quality and credits authority, it credits the right entity — yours.
How to tell if this is happening to you
Entity confusion is easy to miss because it does not announce itself; it shows up as a set of small wrongnesses you might otherwise blame on bad luck. Search your exact brand name and look at what the engine returns: if a knowledge panel appears, are the facts in it actually yours — the right founding year, the right logo, the right description — or are some of them your namesake’s? If no panel appears at all despite a real, established presence, that absence can itself be a sign that the engine is not confident which entity is which. Watch, too, for a brand search that surfaces the other organisation above or alongside you, or for coverage and mentions that never seem to move your standing. Any of these is a hint that your signals are landing on a blurred target.
The confirming test is to look at where your corroboration points. Check whether your Organization markup carries sameAs links at all, and whether the profiles they point to are unmistakably yours and consistent with one another; check that your name, address, and details match across your site, your listings, and your social accounts, rather than drifting into small variations that read as different entities. Inconsistency is what feeds the confusion: every place your details disagree is a place the engine has to guess, and every guess is a chance for it to guess the other entity instead of you. Tightening them — one name, one set of facts, one web of authoritative links — is most of the cure, and it is work you can start today without waiting for anyone’s permission, and it pays off quietly across everything the engine later decides about you.
Entity disambiguation, plainly: quick answers
What does it mean for the engine to confuse my entity with another?
The Knowledge Graph stores the world as entities — organisations, people, places, things — not just as words, and it has to decide which entity a name refers to. When two organisations share a name, it can get that decision wrong: blending your signals with your namesake’s, showing their details in the panel that appears for your brand, or attaching the reputation you built to them. You experience this as a brand that the engine seems not to understand — a wrong or mixed knowledge panel, authority that does not seem to land, mentions that do not add up. The cause is not weak content; it is an unresolved identity.
How do I tell the engine which entity I am?
You give it unambiguous, corroborated signals that point at one identity. The central one is sameAs: link from your Organization markup to the authoritative profiles that are unmistakably you — your verified social accounts, your Wikidata entry, official listings — because each link is a vote that this website and that profile are the same entity. Around it, add the attributes only you have: founding date, location, the specific field you work in, a clear About page, a consistent name, address, and phone everywhere they appear. Third-party mentions that tie your exact name to your site help too. Together these resolve you to one node the engine can hold with confidence.
Why does this matter so much for authority?
Because authority attaches to an entity, and if the engine has the wrong entity, your authority goes to the wrong place. Every signal that should build your standing — a citation, a mention, a strong page — only counts for you if the engine knows it is about you. When your identity is blurred with a namesake, some of that credit leaks to them, and you are quietly competing against a version of yourself. Resolving the confusion does not invent authority you have not earned; it makes sure the authority you did earn is filed under your name rather than scattered across two entities the engine cannot keep apart.
Is this urgent given the core update rolling out now?
It is worth attention, because a core update re-evaluates quality and authority, and that re-evaluation can only help you if it is applied to the right entity. If the engine is unsure which organisation you are, the reassessment lands on a blurred target, and gains you should receive may be diluted or misattributed. Fixing your disambiguation will not, by itself, move you in an update — it is not a ranking trick — but it makes you legible, so that whatever authority you have is seen as yours when the system looks. Think of it as making sure your name is on the door before the inspection, not as a way to pass the inspection.
A note on sources and timing
This is written at the end of May 2022, with the May core update still rolling out. We have described entity disambiguation as it works in Google’s Knowledge Graph: a name resolves to an entity, shared names can be confused, and signals like sameAs, consistent organisation details, and third-party corroboration help the engine hold you as one identity. We have framed it as a legibility task rather than a ranking tactic. The durable point holds regardless of the next update: authority attaches to entities, so making sure the engine knows which entity you are is the groundwork the AC Group has built for 27 years.