Case Study Factory — from "trophy this" to a publishable three-tier case study
Captured win in, publishable case study out — with disclosure rules enforced by machine, not memory.
What was getting in the way.
Even with results captured (Entry #12) and design capability (Entry #5), case studies died in the gap between them: assembling verified numbers, deciding what's safe to disclose, building the page, and QA-ing the claims was a cross-functional project nobody owned. The agency sat on provable wins it never published.
Context
Case-study subjects: e.g., a moving company, a dog-training academy — anonymized or consent-checked per tier.
How the work runs.
- 01
Trigger
a captured, scored portfolio entry (Entry #12) is approved for a case-study build.
- 02
Agent assembles the verified content
Agent assembles the verified content pack: headline stats with source provenance, monthly series for charts, before/after screenshots — no number without a source query may appear at any tier.
- 03
Builds one page rendering three
Builds one page rendering three disclosure tiers from the same content: FULL (absolute numbers), RELATIVE (percentages/multipliers only), REDACTED (fully qualitative, zero digits) — so the same story serves different prospect situations and consent levels.
- 04
Applies the standing disclosure ceiling
Applies the standing disclosure ceiling to every tier (no tenure, no dates, no dollar figures, no source attribution, no itemized service lists).
- 05
Verifies each tier independently on
Verifies each tier independently on the published URL with automated content checks (forbidden strings return zero hits per tier).
- 06
Packages for internal stakeholder review
Packages for internal stakeholder review — review banner, feedback canvas link — before anything is client- or prospect-visible.
- 07
Outcome
a reviewed, disclosure-safe case study ready for sales use at the right tier.
Evidence from the workflow.
Each system has a role.
Intake / record of truth
Case-study catalog + evidence archive
Action
AI site-builder platform
Signal (review)
Visual feedback canvas
Signal (approval)
Team chat
Why this is Coworker.
Takes assignments and reports back. You hand it work, it prepares and returns a result.
Publication is human-approval-gated at two points: entering the build, and going live. The standing disclosure ceiling applies to every tier including FULL.
Effects / Outcomes
Case studies go from "someday" project to a repeatable pipeline: capture → verified pack → three-tier build → automated disclosure QA → review.
Disclosure compliance is machine-checked per tier (the redacted tier is grep-verified to contain no numbers at all) — a category of risk removed structurally.
Clients with little historical data can still get a case study (start at the redacted tier, upgrade as data accrues).
Find where a workflow like this fits.
Start with the systems, work, constraints, and authority already present in your operation.