EXAMPLECO REPORT BRIEF AND CAMPAIGN NOTES
Fictional sample context for an ML Clever tutorial. Not real company results.
Source ID: CONTEXT

N1. Company: ExampleCo, a fictional B2B software business.
N2. Report audience: the Marketing Director and Finance Lead.
N3. Decision: should ExampleCo keep the current channel allocation or test a reallocation within its October 2026 advertising budget?
N4. October's total advertising budget is fixed at USD 20,000. There is no approval to increase it. Any proposed allocation must sum to USD 20,000.
N5. A new paid-search landing page launched on September 15. No randomized experiment was run, and the data does not isolate its effect. Timing alone does not establish that it caused the improvement.
N6. Paid-social targeting changed on September 20. The source pack does not establish why social customer acquisition fell or how the targeting change affected it.
N7. Revenue, margin, retention, lifetime value, and lead quality are unavailable. This report cannot establish return on ad spend, ROI, profitability, or lifetime value by channel.
N8. Finance wants any budget change to begin as a limited, reversible test with a review checkpoint. A recommendation is a proposal, not an approved decision.
N9. Suggested responsible roles are Marketing Director for campaign changes and Finance Lead for budget review. Specific people, dates, thresholds, and approvals have not been assigned; label them as proposed or to confirm.
N10. The report must distinguish observed results, possible explanations, and proposed actions. Do not project new customer counts by assuming September efficiency will continue after a budget change.
N11. Use only this source pack. No industry benchmarks or external research are needed for this example.
N12. Display "Fictional example — sample data" clearly on the report's first page and include a source note on each page with numerical findings.

Source references can use the filename and note ID, for example:
marketing-context.txt, N4-N8.