# Evidence-led Content Brief verification — 2026-08-27

Status: **Verified; review due 2026-09-27.**

## Public synthetic rehearsal

- Topic: `AI agents for marketing`; audience: marketing-operations leaders.
- Keyword validation run `adf4f046-ef62-40d3-8cd7-e3453626c2a9`, final cost
  `$0.017400`.
- Public-source research run `82fb18ff-ef10-470f-8a5f-bab2b3802594`, final
  cost `$0.009800`.
- Total provider spend: **$0.027200**.

The validation returned three useful topic shapes at rehearsal time:

- `AI agents for marketing`: volume 480, difficulty 7, commercial intent;
- `AI agent skills`: volume 720, difficulty 26, informational intent;
- `agentic workflows`: volume 3,600, difficulty 29, informational intent.

The workflow correctly selected the lower-volume commercial phrase as the
primary topic because it best matched the business question. The single source
pass returned five current sources spanning a consultancy operating model,
practitioner analysis, enterprise examples and implementation viewpoints:

- [BCG — agent-native marketing operating model](https://www.bcg.com/publications/2026/agent-native-marketing-operating-model)
- [Ahrefs — agentic marketing](https://ahrefs.com/blog/agentic-marketing/)
- [MarqOps — agentic marketing guide](https://www.marqops.com/blog/agentic-marketing)
- [Moveworks — marketing examples and use cases](https://www.moveworks.com/us/en/resources/blog/agentic-ai-in-marketing-examples-and-use-cases)
- [Ian Kim — agentic marketing blueprint](https://www.iankim.ai/writing/the-agentic-marketing-blueprint)

## Acceptance

The output passed because it produced a specific angle for marketing-operations
leaders, a five-part outline, a source-role ledger and an explicit instruction
not to treat vendor statistics as independent evidence. It stayed under the
trial-safe maximum and recorded all final costs.

This evidence uses a synthetic public topic and contains no customer data,
credentials, approval links or raw provider payloads.