data-centered.com · landscape · graph view

Context-Layer Landscape — Relationships

31 vendors and patterns plotted across the four-axis stack, with the relationships that matter for your strategic position. The dashed anchor in substrate is your moat — producer-time analytics context. Toggle the edge types in the legend to read different threads.

Updated 2026-05-01 Nodes 32 (incl. anchor) Edges Companion view cards →
hover a node to focus its edges

How to read this map

Converges (same primitive)

Indigo dotted lines mark four entries that triangulate the same missing layer: a versioned, lineage-tracked, incrementally-updated corpus. CocoIndex, Cabinet, Cognition/Devin, and Ramp Inspect arrive at it from different angles. The convergence is the strongest evidence that freshness-with-lineage is the analytics-context primitive you should claim.

Plugs in to your stack

Green arrows point at your moat from the OSS / standards / patterns you can adopt without ceding data: CocoIndex (substrate primitive), MCP (delivery standard), Claude Code Tasks (composition), OpenTelemetry (runtime trace), Notion MCP Pattern (delivery pattern). Each is a one-weekend POC away.

Threatens your moat

Red dashed arrows point at your moat from incumbents and aggregators with high data gravity: Snowflake Cortex, Databricks, Salesforce Agentforce, Glean, Atlassian Rovo. None own the analytics-domain context layer yet. The structural difference of producer-time context is the wedge.

Builds on / depends

Faint amber lines show the protocol/standard dependencies inside the stack — almost everything in delivery wraps MCP; Arize contributes to the OpenTelemetry conventions. These are the load-bearing standards; bet on them.

// Tokens linted via DESIGN.md · animation rules via emil-design-eng. Edges hand-curated; not all relationships shown — see node tooltip for what each entry's line means.