Here’s how to structure it:
Two-ad-group strategy
keywords (Exact): Search Match OFF
- Job: own a hand-picked list of competitor and category terms you’ve deliberately chosen.
- Holds: your exact keywords.
- Why it exists: controlled bids on terms you trust, and clean reporting. When one of these converts, you know which term did it.
searchmatch (Discovery): Search Match ON
- Job: let Apple auto-match your app to queries you haven’t thought of, to find new winning terms.
- Holds: no keywords, just the Search Match engine.
- Why it exists: it’s often the actual volume driver early on (the (Low volume terms) bucket is frequently what converts). It surfaces candidates you can later promote into the Exact group.
How they work together
- Discovery finds → Exact owns. When a search term proves itself in searchmatch, you add it as an exact keyword in keywords and bid it deliberately.
- Two guardrails keep them from overlapping:
- Junk negatives at the campaign level, which block irrelevant apps across both groups with one list.
- Your exact keywords as negatives inside searchmatch, which stops Discovery from bidding on terms the Exact group should own, so they don’t compete against each other or blur your reporting.
What actually happens early on
At the start, searchmatch does most of the real work and keywords is the experiment. Your hand-picked list is a guess. It might be a good one, but it takes two or three weeks to find out. Meanwhile Search Match keeps producing installs, so most of your ongoing work is cutting its junk (negatives) and moving its winners into keywords. If the curated terms still haven’t converted after a few weeks, that’s the answer: for your app, Search Match discovery is what works, not hand-picked competitor bids.
The one number you need before any of this means anything is what an install is worth to you. That gives your cost-per-install something to measure against. Pull it from your analytics.