Expertise
Entity and brand SERP
When a machine looks up your name, is it certain which company it found?
What it means
An entity is the thing a machine believes exists behind a name: a company, a person, a product, with attributes and relationships attached to it. Entity work is making that record correct and unambiguous across every source a system consults. The brand SERP is what it looks like from outside, the page of results somebody sees when they search your name, which is often the first and last thing a buyer checks.
Why it matters
Entity resolution happens before anything else. A system that has merged you with a namesake, or attached your string to somebody else's intent, still answers, confidently, with the wrong facts. In search a user gets ten results and corrects for it. An AI answer offers one version and sounds equally certain either way. Every visibility problem downstream of a broken entity stays broken until the entity is fixed.
What you control
- In your hands
- Your structured data, your sameAs links, the consistency of how you describe yourself everywhere you can edit, and the records that can be claimed.
- Not in your hands
- Whether a system accepts your version, how long it takes to update, and which competing entity it happened to learn first.
What we do
- Establish what the major systems currently believe you are, and where their record disagrees with itself.
- Find the collisions: namesakes, former names, and product names that sit inside somebody else's demand.
- Assert one consistent identity across the sources that feed knowledge graphs, and claim the records that can be claimed.
- Connect the people to the company, because author and founder entities are what carry credibility into an AI answer.
Verifiable on this site
PageRadar shares its name with an unrelated software product. Search Console showed the brand term returning a mix of both, with queries meant for the other product arriving here and the brand position sitting well below where it belonged. The organisation record was rebuilt: founders connected as people entities with stable identifiers, sameAs links to the profiles that feed knowledge graphs, and one description used everywhere. The brand term now resolves correctly and the collision queries are gone. View the source of any page here and check the graph.
How we use the framework
Internet Reality is the version of a company the market actually meets. If a system cannot tell which company it is looking at, everything else it publishes about you is a guess wearing confidence.
See the full frameworkCommon questions
Is this just schema markup?
Structured data is how you state the claim. Entity work is making the claim consistent everywhere and getting it accepted, and most of that happens on sources you do not own.
How long does it take?
Weeks to publish, months for systems to accept it. Knowledge graphs update slowly and language models update on their own schedule.
What if a competitor already owns the name?
Then you are competing for an entity rather than for a ranking, and the work is corroboration: getting more independent sources to describe you consistently than describe them.
Start with what the market can see.
A conversation is enough to know whether there is a gap worth closing.