Portfolio Capture — client results and evidence preserved before they vanish
Every win is captured, scored, and provable — before the evidence disappears.
What was getting in the way.
When a client engagement ended, the evidence of the work often ended with it: hosting teardown destroyed the "before" site, praise scrolled away in chat, and results data scattered across five platforms. Great outcomes became unprovable stories within months.
Context
Departing, former, and standout active clients (e.g., a moving company, a plumbing company).
How the work runs.
- 01
Trigger
a team member tags the agent in a client channel: "trophy this."
- 02
Agent races the teardown clock
Agent races the teardown clock first: full-page screenshots of the live site (and archive recovery of the pre-engagement site where DNS already flipped) into the portfolio archive.
- 03
Scrapes everything held on the
Scrapes everything held on the client: memory records, verbatim praise quotes with source references, financial history from local caches, performance stats from connected APIs — each snapshot noting its source query and date range.
- 04
Scores the client as a
Scores the client as a case-study candidate and writes a structured catalog entry.
- 05
Outcome
a complete, sourced evidence pack. Data capture only — page building and publication are a separate approval-gated flow (see Entry #20).
Evidence from the workflow.
Each system has a role.
Intake (before/after evidence)
Client sites + web archive
Signal (lifetime value, tenure)
Payments + CRM caches
Signal (results data)
Ad platform / analytics APIs
Record of truth
Case-study catalog
Why this is Coworker.
Takes assignments and reports back. You hand it work, it prepares and returns a result.
Capture only — never builds a page, never publishes. Publication is a separate human-approval-gated flow.
Effects / Outcomes
Before-shots are preserved ahead of hosting teardown — evidence that was previously lost forever is now archived by default.
Every captured stat carries its source query and date range, so future case studies inherit verified provenance instead of folklore.
Client-level LTV and tenure reads come from local payment/deal caches in seconds with zero API calls.
Find where a workflow like this fits.
Start with the systems, work, constraints, and authority already present in your operation.