Architecture
AuraScore 83/100

Account-Based Real-Time Personalization Engine Target State Blueprint

Design edge-computed, account-based personalization architecture linking identity graphs with sales signals.

Use this prompt when architecting high-performance web personalization and ABM dynamic enablement engines. It produces a target state plan for resolving visitor identities and serving tailored commercial messaging in under 50 milliseconds.

Template

Role: VP of MarTech Architecture & Customer Data Infrastructure Specialist.

Context

  • Identity Resolution Engine: {{identity_resolution_provider}}
  • Ingested Customer Data Sources: {{cdp_data_sources}}
  • Edge Delivery Infrastructure: {{web_edge_runtime}}
  • Sales Enablement Endpoint: {{sales_enablement_integration}}
  • User Consent Framework: {{consent_management_framework}}
  • Multi-Touch Attribution Rules: {{attribution_modeling_spec}}

Task

Design an enterprise target-state architecture and implementation plan for a sub-50ms Account-Based Marketing (ABM) dynamic personalization engine that integrates edge runtime rendering with real-time intent telemetry and sales intelligence dispatch.

Method

  1. Map identity stitching pathways from anonymous IP/cookie tokens into deterministic account tiers using {{identity_resolution_provider}}.
  2. Construct the edge data plane inside {{web_edge_runtime}} for hyper-localized copy and asset injection.
  3. Formulate the synchronization bridge between real-time CDP updates across {{cdp_data_sources}} and the edge key-value cache.
  4. Design the event streaming trigger connecting high-intent site interactions to {{sales_enablement_integration}} in real time.
  5. Integrate {{consent_management_framework}} guardrails to dynamically suppress personalization for opted-out users.
  6. Align downstream session logs with {{attribution_modeling_spec}} without placing telemetry trackers on critical render paths.
  7. Detail an execution plan outlining proofs-of-concept, canary testing, observability baselines, and production cutover.

Constraints

  • MUST restrict total edge decisioning latency to under 50ms at p99.
  • MUST NOT process unconsented or non-anonymized PII across {{web_edge_runtime}}.
  • Fallback to generic baseline copy must occur automatically upon edge cache miss or lookup timeout.
  • Personalization scripts must run out-of-band from primary core web vitals scripts.

Output format

Generate the architecture blueprint structured in 4 core sections:

  1. Edge Personalization Topology & Data Flow (runtime architecture and caching strategy)
  2. Identity Resolution & Consent Matrix (integration details for {{identity_resolution_provider}} and {{consent_management_framework}})
  3. Sales Activation & Telemetry Pipeline (streaming path to {{sales_enablement_integration}} and {{attribution_modeling_spec}})
  4. Execution & Hardening Plan (timeline, load testing criteria, latency SLAs, fallback runbooks)

Self-review

  • Is p99 latency strictly protected under cold-start edge execution scenarios?
  • Are fail-open mechanisms defined for when {{identity_resolution_provider}} experiences degraded service?
  • Does the blueprint satisfy attribution requirements in {{attribution_modeling_spec}} without blocking DOM paints?
AuraScore breakdown
83/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering12/12 · Strong

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness5/5 · Strong

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

developers
developers-architecture
business-strategy-marketing-sales
abm
cdp
edge-computing