App Store Search Ads Explained for Indie Devs (2026)

App Store Search Ads are paid placements at the top of App Store search results, marked with an 'Ad' badge. The bidding model is cost per tap (CPT) — you pay when someone taps your ad to view your product page, whether or not they install. Tap-to-install (CTI) conversion runs 30-60% for well-positioned apps. An auction combines your bid with a relevance score, so bidding alone doesn't guarantee the top spot.

There are two products. Basic is automated and low-control — Apple picks keywords and bids; easier setup, higher CPI. Advanced gives you full control of keywords, bids, audiences, and creative. Almost every indie should use Advanced; Basic is for people who won't learn the system.

Real-world costs in 2026: CPT typically $0.50-5.00 (competitive niches like fitness and dating $2-5; lighter niches $0.50-2; branded keywords often cheapest because you already own the intent). Your real install cost is CPT divided by CTI — a $2 CPT at 40% CTI is a $5 CPI; a $4 CPT at 30% is $13. Then customer acquisition cost is CPI divided by your install-to-paying rate: a $5 CPI at a 5% purchase rate is a $100 cost per paying customer, while a 20% rate (well-onboarded subscription apps) drops it to $25.

Category patterns are worth knowing before you bid. Productivity apps run moderate CPT ($1.50-3) with strong B2B conversion; fitness is high CPT ($2-5) and only works with large LTV; games swing wildly by subgenre; finance carries high CPT but high LTV justifies it; photo and video convert well to free downloads but monetize poorly. The honest math: sustainable Search Ads usage needs either strong subscription LTV or premium pricing. Free, ad-only apps rarely produce positive ROI here. Keep customer acquisition cost at 30-50% of LTV — if LTV is $20 and CAC is $50, the math fails; if LTV is $200 and CAC is $50, it works well. For monetization context, see subscription vs in-app purchases, and for broader reach see how to market an iOS app.

Running everything in one campaign makes optimization impossible — winners and losers blend together. Split your account into 4-6 campaigns by purpose, each with its own match-type approach:

| Campaign | Targets | Match type | CPT range | CTI | Notes | |---|---|---|---|---|---| | Branded | Your own app + brand variations | Exact | $0.50-2 | 60-80% | Cheap, high-converting; blocks competitors stealing your name | | Competitor | Rival app names | Exact | $1-4 | 30-50% | Works when you legitimately compete; disclose clearly | | Generic | Category terms ('habit tracker') | Exact | $1-5 | 20-40% | Higher cost, more discovery, lower conversion | | Discovery | Broad terms to surface new keywords | Broad | low bid | varies | Search-term reports reveal winners to promote | | Custom Product Pages | Audience-specific creative | per source | varies | varies | Different messaging per audience |

Start bids at 50-70% of Apple's suggested figures — their recommendations skew high for indie budgets — and raise only when you're losing auctions on keywords that actually convert. Resist the urge to hike bids on early winners; jumping 50% off thirty conversions usually buys higher costs without proportional volume. Match types matter as much as bids: exact match gives the most control and lowest reach (use it for validated keywords), broad match gives the least control and highest reach (use it only in Discovery campaigns), and Search Match lets Apple choose for you (handy but risks waste, so use it cautiously). Add negative keywords (a budgeting app should exclude 'hotels') to stop broad match drifting into unqualified traffic; a solid negative list takes 1-3 months to build but starts paying off immediately. For ASO context, see the app store ASO guide.

Keyword research decides whether you bid on words that install or waste money on tangents. Start with the reliable freebies — your branded and competitor terms — then brainstorm 50-100 generic keywords across direct descriptors ('expense tracker'), use cases ('split bill calculator'), problems ('how to stop overspending'), and features ('budgeting widget'). Evaluate volume and suggested bids in Apple's dashboard; go deeper with Sensor Tower, Data.ai, or AppTweak ($50-300/month) once you're scaling. Use exact match for proven winners and a separate broad-match Discovery campaign to find new ones, then promote winners.

Don't skip the negative-keyword work mid-research, either: as you brainstorm, note terms that look adjacent but signal the wrong intent, and seed them into your negative lists before launch so broad match never burns budget on them. Building a strong negative list takes 1-3 months of live data, but starting one on day one saves real money.

On the creative side, Custom Product Pages (up to 35 variants) let you match the store listing to the keyword that triggered the ad — show budgeting screenshots to budget searchers, investing screenshots to investing searchers. The first 1-3 screenshots drive most conversion: lead with the benefit, use text overlays, show outcomes not features. App preview videos under 30 seconds beat feature dumps, and the first 170 characters of promotional text are what users see before expanding, so front-load the single most compelling reason to install. Test one variable at a time — a different screenshot lead, then a different video, then different promo text — at least 1,000 impressions per test before you draw a conclusion. Well-run creative lifts CTI 20-50%, which is real money when you pay per tap. The trap: set-and-forget creative slowly decays as users see the same listing repeatedly; refresh quarterly, or sooner when performance starts sliding.

Without tracking you can't optimize. Apple's free Search Ads Attribution API reports impression, tap, install, and post-install events and covers most indie needs entirely. Third-party MMPs (AppsFlyer, Adjust, Singular, Branch) cost $300-3,000/month and only matter once you're running multi-channel at scale. SKAdNetwork is Apple's privacy-first framework — compliant but delayed and aggregated, which slows optimization.

Watch CPT, CTI, CPI, install-to-event rate, cost per paying customer, and ROAS (over 1.0 is profitable, over 2.0 strongly so). Cadence: a daily glance at pacing, a weekly review to pause losers and raise bids on winners, and a monthly strategic pass to test keywords and refresh creative. The ROI-killers are predictable — pausing keywords before they've had 100+ taps, hiking bids 50% off 30 conversions, and neglecting negative-keyword upkeep so campaigns drift toward broad, unqualified traffic. Search Ads is a long game; sustainable performance comes from months of steady optimization, not a clever first setup. For revenue context, see how much do app developers make.

Abstract advice is easy to nod along to and hard to act on, so here's the concrete sequence I'd run with a first $2,000, assuming your unit economics already clear the bar from the previous section. Week one: put roughly $150 into a branded campaign only — your app name and close variations, exact match, a modest bid. That does two cheap jobs at once: it stops competitors buying ads against your name, and it calibrates your tracking so you can confirm the whole pipe (tap, install, post-install event) reports correctly before you spend real money. Almost nobody who skips this step comes out ahead. Weeks two and three: launch one generic campaign with five to eight tightly chosen exact-match keywords at about 60 percent of Apple's suggested bid, plus one low-bid broad-match Discovery campaign whose only purpose is to harvest search terms you'd never have thought of. Budget around $600 across the two weeks and then touch nothing for the first four days — early data lies, and the itch to optimize on day two is exactly where indie budgets get torched. Week four: pull the search-term report from the Discovery campaign, promote the two or three terms that actually converted into their own exact-match ad groups, and dump every irrelevant term you see into a negative list. By now you've spent roughly $900 and you're holding a shortlist of keywords that install rather than a theory about which ones might. The remaining $1,100 goes only into the ad groups showing a cost per paying customer under half your LTV, with bids raised in 15 to 20 percent steps — never the 50 percent leaps that buy cost without volume. And if nothing clears that threshold by the time the $2,000 is gone, the honest read isn't 'spend more' — it's that your listing or your monetization isn't ready for paid traffic yet, and the money is better spent fixing conversion before you buy another tap. Getting that verdict for $2,000 instead of $10,000 is the campaign doing its job.

Search Ads are especially wrong for most indies in their first year of trying to make money from home. Most from-home indies should ride free channels — ASO, content marketing, niche community engagement — until unit economics are proven. Run paid only when you can answer yes to all five: (1) your monetization supports paid acquisition (strong subscription LTV of $50-500+ or $10+ premium pricing); (2) your unit economics are stable and you know your breakeven CAC; (3) your store listing already converts visits; (4) you have 5-10 hours/week for the first three months (dropping to 2-5 after); and (5) you have a $2,000+ testing budget, since productive keyword sets emerge only after $1,000-5,000 of testing.

Skip it when your app relies on viral growth (you'd pay for users who'd come free), when Apple already features you organically, or when you're pre-product-market-fit. Realistic outcomes: branded campaigns are almost always profitable ($1-3 CPI); competitor campaigns often profitable in B2B and subscription categories ($3-12 CPI); generic campaigns are variable ($5-25 CPI); discovery campaigns pay off over months. First profitable campaigns typically emerge after 4-8 weeks. For broader paid strategy, see TikTok ads for beginners, and for ad-spend ground rules Apple's own Apple Search Ads platform documentation.