Change journal and lift

Adwarp keeps an append-only journal of every change made to an ad entity, so you can measure the lift of a change: how performance differed in the window before it versus the window after.

• Where it lives: ~/.adwarp/changelog.jsonl, a plain local file with a JSON object per line. There is no database, no account, and nothing to set up. Override the path with ADWARP_CHANGELOG_PATH.
• Auto-capture: every create or update Adwarp makes to a campaign group, campaign, or creative is journaled automatically. Capture happens at the single HTTP chokepoint every write passes through, so it cannot be forgotten. Ad status changes (adwarp ad pause, update_ad) are recorded under the numeric creative id, the same id adwarp lift creative <id> takes.
• Manual entries: log changes made outside Adwarp (for example in Campaign Manager), or attach a hypothesis, with adwarp changelog add or the log_ad_change tool. Use -l, --label to name the test (for example "outcome-led headline test") and --tags to group related changes.
• Lift: adwarp lift <level> <entityId> (or the compute_lift tool) reads the journal for that entity. For each change, it compares the --window days before (default 14) against the same window after, and reports before and after metrics with per-metric deltas: click-through rate, cost per click, conversion rate, cost per conversion, and more. Recent changes get a clamped, partial after-window.

adwarp changelog add -t creative -i 123456789 -f headline --after "Book a demo in 2 minutes" -l "outcome-led headline test"
adwarp changelog list -i 123456789
adwarp lift creative 123456789 -w 14

This is a directional before and after comparison, not a controlled experiment. It is confounded by seasonality, the LinkedIn learning phase after an edit, and any concurrent budget change, so treat the deltas as a signal.

Set ADWARP_NO_CHANGELOG=1 to turn off auto-capture. Journaling is also off on hosted deploys, which have a read-only filesystem.