Merchandising Content Studio
Outcome
Give merchandisers a governed studio that drafts channel-ready product copy and campaign assets from validated attributes, brand rules, and live performance signals, with brand and legal approval before publish.
| Portfolio attribute | Value |
|---|---|
| Primary persona | Merchandiser |
| Primary aggregate | ContentWorkItem |
| Cosmos DB containers | 2 |
| Grounding data | product data, brand rules, campaigns, performance |
| Development estimate | USD 60-220 monthly dev estimate |
| Runtime options | Python FastAPI and C# ASP.NET Core |
Productivity hypothesis
These measures reflect the workflow, evidence, decision boundary, and industry risk for this use case.
| Impact type | Measure | Conservative internal assumption | Customer measurement |
|---|---|---|---|
| Productivity | Brief-to-approved-draft time | 10-18% reduction | (baseline median draft-cycle minutes - assisted median minutes) / baseline median minutes * 100 |
| Quality or risk | Content rework | 5-10% reduction | (baseline corrected or repeated content-task rate - assisted rate) / baseline rate * 100 |
Important: Conservative internal pilot assumptions only; not public benchmarks, commitments, or expected customer results. Validate against a comparable baseline or matched control cohort with mandatory human review.
Rationale: Merchandisers can create grounded product content faster with fewer avoidable brand and factual corrections.
Pilot guardrail: Keep output editable and require compliance with current product facts, brand rules, and policies.
Customer validation plan
- Baseline the current workflow with representative users and tasks.
- Run the accelerator on the same task mix with human review retained.
- Measure brief-to-approved-draft time and content rework with the documented formulas.
- Replace internal assumptions with customer-observed results before a business case.
Why Azure Cosmos DB
This AI solution needs an operational data plane, not only a vector index:
- Hybrid vector and full-text retrieval combines semantic recall with exact identifiers, codes, names, and domain language.
- Hierarchical partition keys and metadata predicates keep tenant and domain boundaries in the normal query path.
- Operational JSON, source chunks, workflow and agent state, feedback, and audit records remain together for low-latency reads and governed writes.
- Change feed reacts to source and operational updates without polling.
- Autoscale throughput, configurable consistency, and multi-region distribution support variable demand and enterprise availability.
Domain workflow

| Stage | Actor | Input → output | Approval |
|---|---|---|---|
| 1. Select style set and channel targets | Merchandiser | Category assortment and launch calendar → ContentWorkItem with channel targets | Policy controlled |
| 2. Assemble validated attributes and brand rules | Grounding service | PIM snapshot and rule library → Constrained grounding bundle | Policy controlled |
| 3. Generate channel copy variants | Content generation agent | Grounding bundle and performance signals → CopyDraft per channel with claim citations | Policy controlled |
| 4. Run brand, claim, and accessibility checks | Compliance checker | CopyDraft set → Rule check report with blocking findings | Policy controlled |
| 5. Approve and schedule publish | Brand and legal reviewer | Checked drafts and diff against live copy → Approved copy released to channel feeds | Human required |
Technical architecture

Cosmos DB data model
| Container | Hierarchical partition key | Documents | Search paths |
|---|---|---|---|
contentWorkItems | /tenantId, /categoryId | ContentWorkItem, ProductAttributeSet, CopyDraft, PerformanceSignal | /attributeSummaryVectorEmbedding, /copyVectorEmbedding, /attributeSummaryVector, /copyVector |
brandGovernanceRules | /tenantId, /ruleDomain | BrandRule | /ruleTextVectorEmbedding, /ruleTextVector |
Query contract
| Query | Container | Scope | Purpose |
|---|---|---|---|
| Category work queue | contentWorkItems | single-partition | Show every style awaiting drafting or review inside one merchandising category. |
| Draft version history | contentWorkItems | single-partition | Compare successive copy versions for one style and channel during review. |
| Blocking rule violations | contentWorkItems | single-partition | List drafts held by severity-high brand or claim violations before a launch date. The offending ruleId, ruleDomain, and severity are denormalized onto ruleCheckResults when the check runs, so this reviewer query never joins to the brandGovernanceRules container. |
| Underperforming copy sweep | contentWorkItems | single-partition | Find live styles whose conversion fell while returns rose, as refresh candidates. |
Python and C# services route each document type to its declared container and hierarchical partition key. ETag-based writes protect mutable aggregates, while append-only records and bounded cross-partition queries preserve scale.
Azure deployment

Clone the source repository, authenticate with Azure, and run:
Set-Location accelerators/retail-e-commerce/merchandising-content-studio
az login
./Deploy.ps1 -Runtime python -Location eastus2
Change python to csharp to deploy the .NET implementation.
Explore the implementation
- Launch signed-in demo
- Implementation article
- Domain data model
- Workflow and API contract
- Python application
- C# application
- Domain container Bicep
- Platform Bicep
- Cost and configuration
- Source repository