Machine Relations defines entity mass as a descriptive inventory of machine-legible evidence associated with an entity across a declared observation set. It reports the amount and distribution of observed evidence while keeping citation, ranking, trust, and recommendation outcomes separate.
Machine Relations defines Entity Mass as a descriptive inventory of machine-legible evidence associated with an entity across a declared observation set. The inventory can include independent references, attributed claims, structured entity records, and observed citations.
Entity Mass is not a hidden score read from an AI engine. It is an MR measurement construct for organizing evidence that can be inspected directly. A report must declare the sources, query set, engines, observation window, and counting rules used to produce it.
Premise: the construct is useful because one citation does not describe the full evidence environment around an entity. Counting distinct evidence records, their distribution across domains, and their appearance in observed answers creates a reviewable description of that environment. It does not establish why an engine produced an answer.
An Entity Mass record should preserve observable fields rather than inferred engine behavior:
| Field | What is recorded |
|---|---|
| Entity | The entity name and identifier used for the observation |
| Source | The URL and root domain containing the evidence |
| Evidence role | Reference, attributed claim, structured record, or presented citation |
| Claim | The exact statement associated with the entity, when applicable |
| Publication state | Publisher, publication date, and retrieval date when available |
| Observation scope | Query, engine, answer, and observation time when the evidence was observed in an AI response |
| Support status | Whether the cited page supports the adjacent claim when a claim-source receipt exists |
Deduplicate repeated URL records across exports. Reports should state their deduplication rule and retain the underlying records for reproduction.
Working model (not measured): Machine Relations groups the inventory into four dimensions for analysis. These dimensions organize observations; they are not documented ranking factors or universal engine weights.
Treat these dimensions as descriptive. Observe and report citation results separately.
Premise: an Entity Mass report should be calculated only inside a declared scope.
Every Entity Mass report must declare its own scope and counting rules. Compare reports only when those contracts are compatible.
| Concept | What it records | Relationship to Entity Mass |
|---|---|---|
| Entity Resolution Rate | Whether observed answers identify the intended entity under a declared protocol | A separate outcome that can be reported beside the evidence inventory |
| Entity Chain | A documented set of cross-domain references connecting an entity and its claims | A structure for organizing records included in an Entity Mass inventory |
| Domain Authority | A vendor-defined website metric | Not an Entity Mass input unless a report explicitly includes and labels it |
| Share of Citation | Citation frequency within a declared competitive query set | A separate observed citation metric |
| Earned Authority | An MR classification for independent editorial evidence | An evidence role available to the inventory |
Entity Mass describes an evidence set. Entity Resolution Rate and Share of Citation describe observed answer outcomes. Keeping those measurements separate prevents an evidence count from being presented as proof of citation causality.
It is not brand awareness. The inventory contains machine-legible records collected under a stated protocol; it does not measure human recognition.
It is not source quality. A record can exist without being accurate, independent, or supportive of a claim. Those properties require separate classification and claim-source evaluation.
It is not a ranking factor. The construct does not reveal an engine's private retrieval or generation rules.
It is not a business-outcome metric. It does not measure traffic, conversions, revenue, buyer preference, or recommendation likelihood.
It is not causal proof. Inventory changes and citation changes are separate observations.
Premise: Machine Relations uses Entity Mass in the Measurement layer as a structured inventory of evidence available for analysis. The MR Stack can classify where each record originated, but the inventory does not certify that any upstream activity produced a downstream citation.
The practical sequence is: collect evidence, classify it, observe answer behavior, evaluate claim support, and compare compatible observation windows. Limit conclusions to the retained records.
How do you measure Entity Mass? Define the entity, evidence roles, collection method, deduplication rule, and observation window. Count the resulting records by domain and role, then publish the underlying scope with the result. Report citations and claim support separately.
Can Entity Mass be built quickly? No universal timeline is defined. A report can measure how the inventory changed between two declared windows; it cannot infer a standard accumulation rate from the construct itself.
Does Entity Mass replace SEO? No. Entity Mass is an MR evidence inventory. Search ranking, AI citation presence, traffic, and claim support are separate observations.
How should competing entities be compared? The protocol can compare qualifying record totals within the same declared scope. That count remains separate from observed citations and recommendations.
Is Entity Mass the same for every query? No universal value is implied. When query observations are included, results should be segmented by query set, engine, and observation window instead of collapsed into an undocumented global score.
Machine Relations' own methodology, dataset, and research pages related to this term. These are self-references, listed separately from Sources — they are not independent evidence.
Supporting research
Framework context