Three Ways AEP-Heavy Enterprises Respond When the Data Warehouse Outgrows the CDP

Angle

A predictable inflection point is appearing across large enterprises: AEP as the incumbent CDP, a growing data warehouse, data scientists who run everything in Snowflake rather than AEP Query Service, and audiences being computed twice — once in AEP for activation and once in the CDW for ML. The article maps three organizational responses (extend AEP with add-on capacity, supplement it with the CDW as compute backbone while keeping AEP for activation, or replace it with a composable stack) and the organizational and technical signals that determine which path is appropriate. The goal is not to advocate a direction but to name the decision variables that actually differentiate the three paths.

Key decision this helps with

When an enterprise hits AEP's query and export limits, what are the three available architectural responses, and what signals indicate which one to pursue?

Tradeoffs the article will map

• Extending AEP capacity (add-on licensing, predictable vendor relationship) vs. supplementing it with CDW-as-compute (relief from data science friction, but adds reverse-ETL operational surface) vs. replacing it (maximum flexibility, highest political and migration cost)
• AEP-as-edge-node pattern (CDW computes and pushes audiences to AEP for activation via reverse ETL) vs. AEP-as-primary with workarounds (SQL pass-through, data exports) — each avoids a full migration but introduces different operational debt
• Organizational cost of changing the incumbent CDP (contract, retraining, political capital) vs. operational debt of running a split architecture indefinitely

Open questions / uncertainties

• AEP add-on pricing tiers and contract flexibility vary by customer agreement and are not publicly documented — the article should direct readers to their Adobe account team rather than citing specific pricing
• The reverse-ETL approach (AEP-as-edge-node) introduces a new tool (Hightouch, Census, or similar) with its own governance, monitoring, and vendor-relationship requirements — sometimes a hidden complexity in the 'easy middle path'

Knowledge-graph nodes this draws from

• vendor.adobe-experience-platform
• vendor.snowflake
• vendor.databricks
• vendor.bigquery
• concept.data-liquidity
• concept.roi-uncertainty