Should You Use Apple Search Ads Discovery Campaigns?
By Sam H
Discovery campaigns sound like free keyword research. Apple finds searches you did not add, you get installs, you harvest the winners. On a small budget, that pitch often turns into spend you cannot explain and a search terms report you cannot finish.
Discovery is not bad. It is a later tool. This post covers what discovery campaigns are for, when beginners should skip them, and how to run a capped test once Exact learning is already working.
What a discovery campaign actually does
In Apple Search Ads, discovery usually means letting Apple expand beyond your manual Exact list: Search Match, Broad match, or a dedicated discovery-style setup whose job is to surface new queries.
The point is not “more installs.” The point is a harvest loop: new search terms appear, you promote converters to Exact, you negative junk, you keep the discovery budget small so it cannot eat the account.
If you never open the search terms report, discovery is just uncontrolled spend with a nicer name.
Why beginners should wait
- You do not know what “good” looks like yet. Without Exact brand and long-tail baselines, every discovery query looks equally mysterious.
- Thin budgets get diluted. Discovery competes with the keywords you are still trying to measure. See daily budget and how many keywords you should run.
- Search Match on by default already creates chaos. Turning on more expansion before you clean that up doubles the fog. Read Search Match is eating your budget.
- Funnel leaks get louder. More low-intent taps into a weak paywall just accelerate the loss. See CPI looks great, paywall conversion is terrible.
When discovery becomes useful
Turn it on only when most of these are true:
- Brand Exact is stable and category Exact has a few clear winners or losers
- Attribution works and you can see trials or revenue by keyword
- You will review search terms weekly and actually promote or negative
- You can give discovery its own small cap without starving Exact
That usually means after four to six weeks of Exact learning, not week one. Match-type foundations are in Exact vs Broad match.
How to run a clean discovery test
- Separate campaign or ad group. Do not mix discovery into brand Exact.
- Hard daily cap. Treat it as research tuition, often a fraction of your Exact budget.
- One expansion method at a time. Broad or Search Match-style discovery, not every switch flipped.
- Weekly harvest. Winners become Exact. Waste becomes negatives. The middle stays on watch.
- Judge on trials and revenue, not CPI. Discovery can look cheap on installs and terrible on paid conversion. See CPT vs CPI.
What success looks like
A good discovery setup produces a short list of new Exact keywords you would not have guessed, without blowing the learning budget. A bad one produces a long search terms table, flat ROAS, and the urge to “add more keywords” instead of harvesting.
If discovery never yields a promoter after a fair window, pause it. Exact winners already in the account deserve the money more than another week of vague expansion.
Connect RevenueCat before you scale either side. Setup help: Apple Search Ads attribution setup guide. ROAS math: calculate ASA ROAS with RevenueCat. AppSkale puts spend next to subscription revenue so discovery harvest decisions are based on paying users.
Where to go next
Discovery campaigns are optional. Exact brand and long-tail category come first. Add discovery later as a capped research engine with a weekly search-terms harvest, or skip it until the account can afford noise.
Start with Apple Search Ads for beginners and where your first $500 should go. When you are ready to see which discovered terms pay, AppSkale connects ASA spend to RevenueCat revenue at the keyword level.