57-next-ads-plan-global

Read the previous period's report and audit first, then classify campaigns into keep, stop, or fix with numeric evidence. Scale winners by 20-30%, cap new hypotheses at 2-3, split budget 50/30/15/5 across scaling, testing, retargeting, and lookalike, and prepare three budget scenarios.

You need the previous period performance report with spend, CPA, ROAS, and conversions by channel and creative, plus audit findings if problems existed, the new period KPI target, and the budget with any new offers. Missing the report means generating it first.

Test a maximum of 2-3 new hypotheses per period. More than that prevents any single test from reaching statistical significance. Each hypothesis needs one variable, a defined metric, and at least a 7-day timeline.

Stop an ad set when CPA exceeds 2x target for 3 days. Fix it when the diagnosis points to a specific cause: weak CTR means new creative, good CTR with weak CVR means landing page or offer changes, and high frequency with declining CTR means creative refresh.

Yes. The plan adjusts for documented seasonal CPM patterns: Q4 runs about 26% above Q1, November peaks around 27% above average, and January is roughly 21% below. Plans crossing into Q4 should treat the downside scenario as the default assumption.

Use a full media plan when the new period involves a major change such as a new channel, new offer, or new market, or when no prior data exists. The next-period plan is designed for incremental cycles built on existing performance data.