analysis-campaign

Post-experiment analysis is structured through a durable campaign charter that governs multiple analysis slices, including ablations and robustness checks, providing per-slice run artifacts and aggregated reports aligned to a research plan.

A campaign charter is a structured plan and checklist that coordinates follow-up experiments to test main claims. It provides a traceable evidence path with durable artifacts like PLAN.md and CHECKLIST.md to manage multiple analysis slices.

Baselines and evidence are managed through a writing-facing workflow integration that binds analysis slices to manuscript sections. This produces evidence-led narratives with traceable per-slice run artifacts and aggregated campaign reports.

Yes, the structured analysis workflow applies to both writing-facing campaigns and internal follow-up runs such as error analyses and boundary investigations, providing durable per-slice artifacts and aggregated reports for any coordinated evidence campaign.

The best way to start a coordinated evidence campaign is to define the campaign objective and main claim, identify the most critical follow-up questions, and launch the first analysis slice to address those specific questions.

Existing research outlines are beneficial but not strictly required, as the campaign framework provides durable artifacts and aggregated reports that can align with a selected research plan or manuscript outline to produce evidence-led narratives.