Surge V3.6 Framework — System Architecture
The Surge Architecture Stack
Purpose
This artifact exists to show how Surge's components stack — from foundational datasets through asset production, node deployment, trust targeting, and surface visibility — so practitioners understand where each phase fits in the architecture.
How to Read This Stack
Bottom-up: Each layer depends on the layers below it. You cannot build Asset Packs without understanding Datasets. You cannot deploy Nodes without Asset Packs. Trust Neighborhood targeting requires a Node strategy. Surfaces are the outcome layer — where visibility becomes measurable.
Top-down: If something isn't appearing on a Surface, trace downward to find the gap.
Top-down: If something isn't appearing on a Surface, trace downward to find the gap.
LAYER 5
Surfaces
"Where do we need to appear?"
Maps (GBP)
Web (Organic)
Video
Images
Audio/Podcast
AI Features
LAYER 4
Trust Neighborhood
"Who should validate us?"
S0 Seeds
S1 Hubs
S2 Validators
S3+ Standard
Vertical Mapping
LAYER 3
Node Tiers
"Where do assets get deployed?"
Primary Nodes (Controlled)
Secondary Nodes (Syndicated)
Tertiary Nodes (Automated)
LAYER 2
Asset Pack Engine
"What gets produced for each intent?"
Primary Page
GBP Post
Short Video
Audio Clip
Images
Schema Plan
LAYER 1
8-Dataset Model
"What types of information patterns are needed?"
Correlative
Amplifying
Collaborative
Parallel
Inverse
Bridging
Conflicting
Emergent
LAYER 0
Entity Fact Sheet
"What is canonically true about this entity?"
Canonical Name
Entity Type
Primary URLs
Service List
Geo/Service Areas
Proof Assets
Dependency Rule
No layer can produce reliable outputs if the layers below it are incomplete or inconsistent. An Asset Pack built without a locked Entity Fact Sheet will propagate errors. Nodes deployed without a Dataset strategy will have coverage gaps. Trust targeting without Node architecture is random distribution.
Layer Specifications
Entity Fact Sheet
Foundation Layer
Function
Single source of truth. All content, schema, and distribution must align with these canonical facts.
Critical Rule
No publishing until locked. No scale until verified. All AI drafts checked against this document.
Failure Mode
Contradictions across sources. NAP inconsistency. Invented claims that can't be verified.
8-Dataset Model
Conceptual Layer
Function
Classifies how information patterns reinforce machine confidence. Determines coverage strategy.
Critical Rule
Coverage across types matters more than volume within one type. Score each dataset per intent pack.
Failure Mode
Over-indexing on one dataset type (e.g., all Correlative, no Collaborative). Unbalanced coverage.
Asset Pack Engine
Production Layer
Function
Repeatable production unit. One intent pack = one Asset Pack = multi-format content with consistent messaging.
Critical Rule
Titles and metadata must match the intent phrase across all assets. Human QA before publish.
Failure Mode
One-off content creation. Inconsistent naming. Publishing without Entity Fact Sheet verification.
Node Tiers
Deployment Layer
Function
Defines WHERE assets go. Primary (controlled), Secondary (syndicated corroboration), Tertiary (automated amplifiers).
Critical Rule
Deploy Primary first, then Secondary, then Tertiary. Each tier links back to canonical sources.
Failure Mode
Tertiary nodes publishing unreviewed content. Broken link chains. Secondary nodes with wrong canonical facts.
Trust Neighborhood
Distribution Layer
Function
Prioritizes distribution targets by trust distance. S0 seeds → S1 hubs → S2 validators → S3+ standard platforms.
Critical Rule
Pursue S1 mentions first. Map the vertical's trust core before distributing randomly.
Failure Mode
Random distribution without trust mapping. Pursuing S3+ platforms while ignoring S1 hubs.
Surfaces
Outcome Layer
Function
Where visibility becomes measurable. Maps, Web, Video, Images, Audio, AI Features.
Critical Rule
Measure visibility levels (0-3) per surface per query cluster. Trace gaps downward to find root cause.
Failure Mode
Measuring only one surface (e.g., Web) while ignoring others. No baseline screenshots for proof.