Apple Ads Basic or Advanced: Which One Should You Run?

For years mobile teams called this channel Apple Search Ads. Apple renamed the surface Apple Ads and now sells two separate things under it (ads.apple.com, retrieved 2026-09-01). Growth teams still argue about in-house versus retainer, priced as labor. The tier decides how much work exists in the first place, and in Basic most of that work is unavailable to you at any budget.

Source: Apple, compare Apple Ads solutions, retrieved 2026-09-01.

Basic stops at $10,000 a month per app

The cap is not a suggested budget. It is the maximum Apple will let one app spend in Basic in a month (Apple). The month your plan calls for more, the tier is the constraint, and the migration happens whether you scheduled it or not.

The second difference costs more and arrives later. Basic does not hand back keyword-level search term data, because you never set keywords in it (ads.apple.com/app-store/basic). AppTweak’s tier comparison reads the same way (apptweak.com). So the account cannot tell you which query bought which install.

Anyone who has run Basic knows that feeling already. Spend is visible. Installs are visible. What people typed is not. Move up to Advanced and the keyword build starts from nothing.

The control Advanced sells you is the keyword

In Advanced you set the keywords and a bid against each one, and you write the negatives (Apple). That is the whole difference in daily work.

Setting your own bid earns its place under one condition. You need enough taps on a keyword each week that a bid change produces a readable difference in cost and in trials. Below that you are moving a number and reading noise.

The condition holds per keyword, not per account. A $50k month spread across hundreds of terms can still be too thin to bid by hand on any single one, on our own arithmetic.

What each of the four placements buys you

Advanced serves the Today tab, the search tab, search results, and product pages (ads.apple.com). Three of those four are reach. Somebody opened the App Store, or scrolled a competitor’s listing, and your ad appeared.

Search results is the placement where a person typed something. There a keyword means what you think it means, and the search term record it produces is what you carry into every later bid.

Custom product pages are the variable most advertisers leave on default. Apple lets a custom product page serve as the creative for Advanced search results ads (Apple), so the screenshots a searcher sees can match the theme they searched. Ship one page for every intent and you pay auction prices for a mismatch you chose.

Getting the tier wrong costs a quarter of unreadable installs

Apple advertises conversion rates over 60% for ads at the top of search (ads.apple.com). That gets quoted as a selling point. Read it as a warning about your reporting. When almost every tap installs, install cost stops separating good queries from bad ones, and Basic bills you on exactly that metric.

Stage the arithmetic on a hypothetical account so the loss is visible. Two installs land the same week, one at $4 and one at $9. The $4 install never reaches a paywall view. The $9 install starts a trial and renews. In Basic both collapse into one undifferentiated line item, and nothing in the report says which query to buy more of. Three months at the ceiling is $30k of spend, by our own arithmetic, with no keyword-level record of what it bought.

The read is still available outside the ad account. Your measurement partner can join installs to trial starts and renewals, and the plumbing choices are in our Adjust versus AppsFlyer comparison. Basic will not itemize any of it by keyword. Grade the channel on days to payback at 30, 60, 90 or 180, using the definitions in our mobile app KPI doctrine.

We do not report cost per install as the grade for an Apple Ads account. It is a bid input.

“Apple’s bidding is good now, so why not just run Basic?”

The argument runs like this. Apple handles the bid, Basic asks almost nothing of you, so pick it, save the hours, and get back to the product. Founders say it out loud in board meetings, and it is not a stupid position.

It holds under conditions you can name. One app. One market. Spend under $10,000 a month. No plan to expand the catalog, and no intention of building a keyword asset. Independent tier guides land in the same place for teams that small (apptweak.com).

It stops holding at the cap and at the search term gap. No amount of automation moves the $10,000 ceiling.

What stays manual at any volume, and what we refuse

Three things stay manual whatever your spend. Negative keywords, because discovery is a search term generator and unpruned discovery becomes a tax. Match types, because promoting a paying term to exact with its own bid is how the keyword asset gets built. Product page variants, because no bidding system has read your paywall.

One refusal costs us money to say. We would not open a second storefront before the first one pays. Twenty markets at low spend produce twenty datasets too thin to act on, which is the same volume problem as hand-bidding a keyword with four taps a week.

The two triggers that decide it

Open the search terms report in Advanced and sort by taps in the last week. The rows with enough taps that a bid change would show up in cost and in trials are the only rows worth touching by hand. Everything below them waits for volume.

The other trigger sits in your plan rather than the account. If the plan needs more than $10,000 per app in a month, or a second market, or a keyword report you can act on, the tier is already decided and the extra hours are the price of that decision. Nobody loses a quarter in this channel to a bid they set slightly wrong. They lose it to a Basic account that spent to the ceiling and cannot name a single query that paid.