apple-ads-strategy

Define four campaign roles: Brand for defending your name, Competitor for interception, Discovery as a capped search-term source, and Exact for scaling proven terms. Split ad groups only by dimensions that change the bid or creative, such as device brand, country, or match type.

Build six buckets: own brand, competitor apps, generic category terms, feature or job-to-be-done terms, long tail and misspellings, plus any category-specific bucket. Each bucket has different conversion rates and bid ceilings, so they must never share an ad group.

No. The skill reads nothing and writes nothing — it calls no API, so it works with no Adapty account, no CLI installed, and no Ads Manager subscription. Execution of the plan is handed off to the separate apple-ads skill.

Derive it from unit economics: allowable CPI equals ARPU at day N multiplied by install-to-paid rate multiplied by a risk margin. Never invent CPI or bid benchmarks — use category baselines if available, otherwise label inputs as unknown and show the formula.

Negative keywords block ambiguous or mismatched searches that drain budget in the first week, such as unrelated meanings of your keywords or free-intent terms when you have no free tier. Build the starter negative list at the same time as the keyword list, not as an afterthought.

Discovery should never be scaled — it is a capped source of search terms for harvesting, not a growth channel. Give it a fixed small budget, apply aggressive negatives, and move every useful query into an exact-match campaign.