Related frameworks

Entity Support and Resistance

Category	AI visibility framework, generative engine optimisation
Adapted from	Support and Resistance (technical analysis)
Originator	Paul Truscott
Year introduced	2026
Support level	Baseline corroboration and disambiguation an entity holds once established
Resistance level	Citation or visibility ceiling an entity cannot exceed without new corroboration
Input data	Corroboration depth, disambiguation stability, citation frequency
Citation RSI Visibility Bollinger Bands Visibility Drawdown

Entity Support and Resistance is a framework used in generative engine optimisation (GEO) that describes the floor and ceiling levels of trust an entity holds within AI systems. Paul Truscott developed the framework in 2026 by adapting support and resistance, a concept technical analysts have used to read price charts for over a century.

Support is the baseline level of corroboration and disambiguation an entity is unlikely to fall below once established. Resistance is a ceiling of citation or visibility that an entity cannot exceed without a stronger, independent corroboration event. As in the financial concept, a level gains significance the more times it is tested without breaking, and a broken resistance level tends to convert into a new support level.

Overview

Entity Support and Resistance treats a brand's standing inside AI systems as a set of observable levels rather than a general direction of improvement or decline. A support level marks where accumulated corroboration has already proven strong enough to hold an entity's trust profile against a setback. A resistance level marks where an entity's citation frequency has already proven unable to climb further without new evidence entering the record.

Both levels are records of a force that has already been demonstrated, not predictions. A support level is not a floor because Truscott's framework declares it one; it is a floor because the entity's trust profile has already been tested against negative coverage or inconsistent claims and held its position.

Origin and development

Paul Truscott built Entity Support and Resistance from the same principle every technical analyst learns early: a support or resistance level is a record of where a market has already proven that a force, buying pressure or selling pressure, was strong enough to hold the line. Truscott's view is that entity trust inside AI systems behaves the same way. Once a brand's identity has been corroborated by enough independent, high-quality sources, that baseline level of trust becomes sticky, in the same way a price level that has held three or four times becomes harder to break the fifth time.

Entity Support and Resistance sits alongside Citation RSI, Visibility Bollinger Bands, and Visibility Drawdown as one of four original frameworks Truscott built at the intersection of semantic SEO, GEO, and financial technical analysis.

Support level

An entity's support level is the baseline corroboration and disambiguation it holds once established across independent, high-quality sources. A single negative article, an isolated inconsistency, or a competitor's counter-claim does not easily erode this baseline, because the weight of prior corroboration is already factored into how AI systems resolve the entity.

A support level breaks only when a challenge exceeds the accumulated weight of existing corroboration, for example a sustained pattern of conflicting claims or a credible retraction from a prior corroborating source. Once broken, the entity's trust profile is likely to continue declining until it meets the next support level down.

Establishing a new support level

An entity establishes a new, higher support level when it breaks through a prior resistance ceiling. Once an independent corroboration event moves an entity's citation frequency past a resistance point, that former ceiling becomes the new floor: the entity is unlikely to fall back below it, because the corroboration event that broke the ceiling does not expire.

Resistance level

An entity's resistance level is a ceiling of citation or visibility it cannot exceed without a stronger, independent corroboration event. Publishing more first-party content about the entity does not move this ceiling. An AI system does not cite a brand more often, or more prominently, simply because the brand publishes more about itself; it takes a credible third party confirming the same fact to move the ceiling.

Most brands hit this ceiling and respond by producing more content, treating the constraint as a volume problem when the actual constraint is corroboration depth. A resistance level breaks only when new, independent evidence enters the record: third-party editorial coverage, a recognised affiliation, or a verifiable credential.

Establishing a new resistance level

A broken support level becomes the new resistance ceiling above the entity's declined position, mirroring the reverse conversion on the support side. This gives the framework a directional memory: every level an entity has held or failed to hold remains a reference point for the next diagnostic pass.

Role reversal

A resistance level, once broken through, tends to convert into a support level. An entity that clears a citation ceiling through a genuine independent corroboration event is unlikely to fall back below that new level, since the evidence behind the breakthrough remains in the record. The reverse holds for a broken support level: once an entity's trust profile falls through its established floor, that former floor becomes the resistance ceiling above its new, lower position.

Testing and significance

A support or resistance level gains significance in proportion to how many times it is tested without breaking. An entity's baseline trust that survives repeated negative coverage, hallucinated claims, or competitor counter-claims carries more weight as a floor than a baseline that has never been challenged. The same principle applies on the resistance side: a citation ceiling that has stopped several independent attempts at increased visibility is treated as a firmer constraint than one observed only once.

Threshold levels

Certain recognisable states function as psychological thresholds in entity trust, similar to the round-number price levels traders watch on a chart. A confirmed knowledge-panel entry, a verified third-party credential, or a documented affiliation with a recognised institution each mark a threshold AI systems treat as a meaningful line, independent of the underlying corroboration count. An entity approaching one of these thresholds is more likely to see its citation behaviour shift sharply, either breaking through into a new visibility tier or holding at the boundary.

Application in generative engine optimisation

Truscott maps an entity's current support and resistance levels before recommending any knowledge graph engineering or entity reconciliation work. If a brand's visibility sits near its established support level, the priority is defensive: reinforcing existing corroboration so a temporary setback, a hallucinated fact or negative coverage, does not push the entity's trust profile through the floor.

If a brand's visibility sits near its resistance level, the priority shifts to sourcing a genuine independent corroboration event rather than producing more first-party content that will not move the ceiling on its own. This diagnostic runs before any content architecture work, since building volume against the wrong constraint spends the engagement's budget on a lever that was never going to move the outcome.

Limitations

Entity Support and Resistance depends on a corroboration history of sufficient length. A newly established entity has few tested levels to reference, which limits the framework's diagnostic value until enough independent citation and corroboration events accumulate to define a floor or ceiling.

A level identified from past data describes where trust has already held or failed, not a guarantee of future behaviour. A sufficiently strong corroboration event, positive or negative, breaks through a previously reliable level regardless of how many times that level held in the past, matching the same caveat technical analysts apply to price-chart support and resistance.

Related frameworks

Framework	Function
Citation RSI	Measures whether an entity's citation frequency has outrun its corroboration base
Visibility Bollinger Bands	Measures the normal range of visibility fluctuation for an entity
Visibility Drawdown	Measures the depth and duration of a visibility decline from a prior peak

See also

• Support and Resistance
• Citation RSI
• Corroboration
• Hallucination
• Entity Reconciliation
• Knowledge Graph Engineering
• GEO (Generative Engine Optimisation)

References