Apple Search Ads Keyword Bidding Strategy
Apple Search Ads keyword bidding is one of the few paid acquisition disciplines where deliberate strategy still beats algorithmic automation in 2026. Apple's auction structure rewards operators who structure campaigns intentionally, segment by match type, and manage bids against the data they actually have. The teams that run ASA on default match types and Apple's recommended bids leave 20 to 40% of efficiency on the table.
This article is the bidding playbook Semnexus uses across ASA engagements in 2026. It covers the four-tier campaign structure, match-type allocation, negative keyword discipline, and the bid-management cadence that compounds.
The four-campaign structure
A correctly-structured ASA account has four campaigns per app, segmented by intent and match type:
| Campaign | Match types | Purpose |
|---|---|---|
| Brand defense | Exact + Broad on your brand | Defend branded search; cheapest CPI |
| Category exact | Exact match on category head terms | Highest-intent category traffic |
| Category broad | Broad match on category terms | Discovery; feeds the exact campaigns |
| Competitor exact | Exact match on competitor brands | Selective competitor coverage |
This structure separates traffic by intent, lets you bid differently per intent tier, and surfaces the negative keywords you should be adding to the broader campaigns.
Match type strategy in 2026
ASA supports two match types — Exact and Broad — plus Search Match (automated keyword discovery). Each has a specific role:
Exact match
Use for keywords you've validated as high-converting. Exact match shows your ad only when the user types the exact phrase or a close variant. CPI is highest but conversion is also highest. The bid should be your highest tier.
Broad match
Use for discovery on category terms. Broad expands the user's search to include relevant variations, plurals, and adjacent intent. CPI varies widely. Bid lower than your Exact tier and use this campaign to identify which broad-matched search terms convert; promote them to Exact.
Search Match
The Apple-managed automatic keyword discovery. Useful as a small percentage of the account budget for surfacing new keyword candidates. Should not be a primary source of installs because you cannot control which terms Apple surfaces.
Bid tier framework
Bids should not be set per keyword in isolation. They should be set per tier of conversion behavior:
| Tier | Examples | Bid posture |
|---|---|---|
| Tier 1: Branded exact | Your brand name and close variants | Highest |
| Tier 2: Category head exact | "budget app," "language learning" | High |
| Tier 3: Category long-tail exact | "budget app for couples," "language learning for travel" | Medium-high |
| Tier 4: Category broad | All category terms in Broad match | Medium |
| Tier 5: Competitor exact | Competitor brand names | Low to medium, selective |
A typical tier ratio: Tier 5 bids at 30-50% of Tier 2 bids; Tier 4 at 60-80%; Tier 3 at 80-100%; Tier 1 floors at whatever wins the auction (often very low because competition is weaker).
Negative keyword discipline
Negative keywords are the most underused ASA control. The discipline:
Run negative reports weekly
Pull the search terms report for Broad-match campaigns. Identify terms that brought clicks but no installs, or installs but no activation.
Add irrelevant terms to campaign negatives
Brand misspellings that bring the wrong audience, adjacent categories that bounce, and intent mismatches (informational searches vs install intent).
Move converting Broad terms to Exact
The search terms report also surfaces what's working. Promote them to Tier 2 or 3 Exact at the appropriate bid tier.
Run cross-campaign negatives
Add Brand terms as negatives in the Category campaigns; add Category head terms as negatives in the Competitor campaigns. Prevents cannibalization.
This negative-keyword loop is the single biggest CPI compression lever in ASA, and most accounts run it once a quarter or not at all.
The bid management cadence
Bids should not be touched daily; the auction needs time to stabilize. The right cadence:
Daily
Check spend pacing and any zero-spend keywords. Pause anything obviously broken.
Weekly
Run the negative-keyword sweep. Move Broad winners to Exact. Adjust bids on keywords whose performance has shifted by 30%+ vs the prior period.
Monthly
Rebalance the Tier structure. Review which Tier 3 keywords have earned promotion to Tier 2 (or vice versa). Audit campaign budget caps.
Quarterly
Refresh the keyword universe. Add 20-30 new long-tail candidates from the search terms report and category research. Sunset chronic non-converters.
Touching bids more often than weekly produces auction instability without performance gains.
Five common bidding mistakes
The patterns we see in audits:
- Single-tier bidding. All keywords at the same bid. The account systematically overpays for low-converting terms and underpays for the high-converting ones.
- No negative keywords on Broad campaigns. Broad spend leaks to irrelevant terms.
- Using only Search Match. Cedes too much keyword decisioning to Apple; produces inconsistent CPI.
- Bidding too low on branded. Even at floor bids, branded should always win against a competitor's defensive bid.
- Setting bids to "recommended." Apple's recommendations are anchored to the auction, not to your conversion economics. Use them as a reference, not a target.
Custom Product Pages and bidding
When you run Custom Product Pages (CPPs), each ad group can route to a specific CPP. The right CPP raises conversion rate by 10 to 25%, which means the same bid produces a lower CPA. This compounds with the bidding strategy:
- Tier 2 Category Exact keywords can support a 10-25% higher bid when paired with a matching CPP.
- Tier 5 Competitor Exact keywords often need a higher bid AND a tailored CPP that addresses the competitor comparison.
The full ASA optimization is bid x match type x CPP. Optimizing only one of the three caps the channel.
Measuring success
The minimum scorecard, weekly:
| Metric | Where to find it | Target |
|---|---|---|
| CPI by tier | ASA + MMP | Tier 1 lowest; Tier 5 highest; Tier 2 the workhorse |
| Activation rate by tier | MMP | Tier 1 highest; long-tail should match or exceed Tier 2 |
| Search Terms report coverage | ASA Search Terms | New negatives added each week; new Exact promotions each week |
| Account CPI trend | ASA reporting | Stable or declining over 90 days |
| Total install share by tier | ASA + MMP | Tier 2 and Tier 3 combined should be 50%+ of installs |
Frequently asked questions
Should I run ASA Advanced or ASA Basic in 2026? Advanced unless you have no operator capacity. Basic is a useful placeholder for very early stage; it does not let you optimize keywords or placements.
How many keywords should an ASA account have? For most apps at Stage 1-2, 50-150 unique keywords across the four-campaign structure. Larger accounts can have 300+, but quality of keywords matters more than quantity.
Should I bid on competitor brands aggressively? Selectively. The top 1-3 competitors in your category are often worth defensive coverage. Beyond that, competitor bidding produces high CPI with mediocre conversion.
What about Custom Reports and automated bidding rules? Useful as guardrails (e.g., auto-pause keywords with CPA above $X), not as primary optimization. The auction rewards human judgment more than rules at this stage.
How does ASA bidding interact with ASO? Tightly. Higher organic ranking on a keyword improves the quality of the user clicking your paid ad (they see your brand twice). Coordinated ASO + ASA on the same Tier 2 keywords produces noticeable CPI compression.
If your ASA account is on default bids or you suspect your tier structure is leaking efficiency, the Semnexus mobile app marketing team runs ASA audits and bid restructuring as part of every paid engagement. The website marketing team handles cases where ASA bidding is coordinated with broader category-level SEO.
