Brand vs Non‑Brand Keywords in Apple Search Ads: a Practical Split to Defend Downloads Without Overpaying

Indie teams often treat “keywords” as one bucket in Apple Search Ads. The problem: brand searches (people already trying to find your app) behave very differently from non-brand searches (people browsing category demand). If you mix them, your reporting gets hard to interpret and your bids drift upward for users you didn’t really need to “buy.”

This post gives you a concrete way to split brand vs non-brand keywords into separate ad groups/campaigns, bid them differently, and decide when brand defense is costing you money.

Why brand and non‑brand are not the same lever

On Search Results, both brand and non-brand keywords compete in the same CPT auction model. But their downstream behavior differs:

Brand intent usually means higher TTR (more taps per impression) and often higher conversion rate (more likely to install because the user is already looking for you).
Non‑brand intent tends to have lower TTR and lower conversion, but it’s where you generate incremental demand.

Because of that, a single blended campaign can hide the truth. Your overall ROAS may look “okay,” but brand might be inflating efficiency while non-brand quietly burns budget—or vice versa.

Step 1: Define what “brand” means for your app (and be strict)

Start by creating a short, explicit list. A good brand set includes:

Your app name (exact spelling, common casing)
Your developer name (publisher name shown in the App Store)
Distinctive brand tokens from your branding (only the ones users actually type)

Avoid accidentally treating category phrases as brand. If “Acme Notes” is your app and “notes” is too generic, keep it out of the brand bucket.

Tip: use Search term harvesting to validate

Even without changing anything yet, you want to confirm what people really search. In ASA, look at search terms surfaced to your app for a week (or two, depending on spend). If you see junk matching your app name due to broad keywords, tighten it.

Step 2: Put brand and non‑brand into separate ad groups (at minimum)

You can do this with ad groups inside one campaign, but I strongly prefer separate campaigns when possible.

Recommended minimum structure

Campaign A (Brand Defense)
- Ad Group 1: Brand — Exact
- Ad Group 2: Brand — Broad (optional; usually only if you have low waste)
Campaign B (Non‑Brand Acquisition)
- Ad Group 1: Non‑brand — Exact
- Ad Group 2: Non‑brand — Broad (or Discovery/Search Match, depending on your control level)

Why ad groups? It lets you bid and evaluate separately. Brand typically deserves a different CPT ceiling than non-brand.

Step 3: Bid brand defensively, but don’t “train” the auction with unlimited budget

Brand defense is about keeping your app visible when users search for it. You don’t need to pay “leaderboard prices” if your TTR is already strong.

Here’s a practical way to set initial bids:

Start from your current brand keyword performance (or a conservative estimate if you’re new).
Compare CPT and TTR:
- If brand keywords already have high TTR, you can usually cap bids lower than non-brand without losing too many taps.
If you use Broad for brand, be careful:
- Broad can match user queries that aren’t actually looking for your app.
- If your “brand broad” includes irrelevant queries, it will quietly raise your blended CPT.

A simple rule of thumb

Brand exact: bid for stable visibility and let conversion do the work.
Brand broad (if you keep it): bid like it’s a test, not like it’s guaranteed intent.

If you want one concrete process: run brand exact at a steady bid, and treat brand broad as a small experiment. If it can’t maintain good conversion rate relative to its CPT, remove or restrict it.

Step 4: Use non‑brand to find incremental users—then graduate winners

Non‑brand is where your auction exploration matters.

For non-brand keywords, focus on identifying which match types are actually producing incremental installs:

In non-brand Exact, you get tighter query intent. If Exact is decent, you can scale cautiously.
In non-brand Broad, you’ll often discover search terms you didn’t think of. But broad will also produce irrelevant taps.

To manage that, apply a feedback loop:

Watch TTR and conversion rate (installs/taps).
If taps come but installs don’t, your issue is often either:
- keyword intent mismatch (wrong audience), or
- store conversion is too low (product page / screenshots / pricing model / age gating).

This isn’t an excuse to ignore bids—just know which lever is more likely.

Step 5: Evaluate brand with “incrementality mindset,” not vanity ROAS

Apple Search Ads gives you performance metrics, but it won’t tell you whether an install would have happened without ASA exposure.

So when evaluating brand, ask:

Is brand improving overall efficiency? Sure—but that might not mean it’s driving incremental downloads.
Are brand campaigns still efficient when you isolate them? That tells you if your bids are reasonable.
Does non-brand improve after you cap brand bids? If yes, you likely freed budget from non-incremental spend.

What to look at in the numbers

For each bucket (Brand vs Non‑Brand), track:

CPT (are you paying too much to “buy” certainty?)
TTR (are impressions turning into taps?)
conversion rate (installs/taps)
CPA/CPI and ROAS

If brand shows stellar ROAS but non‑brand is weak, you may be overspending on defense while acquisition needs store or keyword work.

Step 6: Don’t let brand performance hide non‑brand problems

Once you separate reporting, take action based on imbalances.

If Brand is efficient but Non‑Brand is not

Reduce brand broad first (biggest risk of irrelevant matching).
Lower brand bids slightly if they’re far above what’s needed for stable visibility.
Move that saved budget into non‑brand exact winners or new keyword tests.

If Non‑Brand is efficient but Brand is expensive

Lower brand exact bids to a defensive ceiling.
Consider removing brand broad entirely.
Keep brand only to the extent it’s still returning acceptable CPI.

If both are inefficient

The issue might not be bidding.
Check whether your app page (and any Custom Product Pages you use) matches the intent of those keywords.

(There’s no creative auction edge here—keywords, bids, targeting, and the store page are the real levers.)

Step 7: Make scaling decisions using bucket-specific guardrails

When it’s time to scale, scaling brand and scaling non‑brand should not be identical.

Brand scaling guardrail

Increase bids or budgets only until you see CPI degrade meaningfully.
If brand installs are “cheap,” scaling can still be fine—just remember you’re mostly purchasing reach among users who already want your app.

Non‑brand scaling guardrail

Scale only after non-brand shows consistent conversion rate and acceptable CPA/ROAS.
Promote strong non-brand search terms from discovery/harvest into Exact ad groups where you have control.

Common mistakes to avoid

Mixing brand and non‑brand in the same ad group → you lose the ability to tell which intent is causing your results.
Keeping brand broad without reviewing search terms → brand broad often contains surprising non-brand queries.
Treating overall ROAS as one number → it’s easy to hide non-brand inefficiency behind brand efficiency.
Scaling bids purely because brand metrics look good → brand can look “easy,” but it may not be where incremental value is.

How AdsBuddy fits (lightly)

If you’re already separating brand and non‑brand, you’re doing most of the hard work. Tools like AdsBuddy can then help by reading your ASA metrics plus your revenue signal and producing a short, prioritized list of daily changes—so you can quickly see whether brand defense is costing more than it should, or whether non‑brand needs re-keywording or store fixes. You still approve every change; it’s advisory, not autopilot.

Takeaway

Treat brand and non‑brand as two different businesses inside Apple Search Ads:

Brand defense buys visibility for people already looking for you—bid defensively and keep broad under control.
Non‑brand acquisition buys incremental demand—measure it separately, promote winners into Exact, and only scale when it holds up on conversion.

When you isolate intent, your ROAS story stops being confusing—and your next bid change stops being a guess.