A First Look At Apple’s App Store Search Ads


Update: please see our post-launch guide to Apple Search Ads for a detailed look at the Apple Search Ads platform after it went live in October 2016.

Last week, Apple unveiled plans to launch a search advertising platform for the App Store. This announcement was made in anticipation of WWDC 2016, which kicked into gear with a two-hour keynote that included an extensive introduction to the platform. The launch of Apple Search Ads is a big deal for app developers as it marks the emergence of a very promising user acquisition channel, and we want to share some early thoughts on the platform.

Apple Search Ads shares many core characteristics with Google AdWords, so it should feel very familiar to paid search marketers. But there are also some very important differences that shouldn’t be overlooked. Below we’ll walk through the similarities and differences and share our initial insights on how Search Ads are going to impact mobile app developers.

What’s Similar


Campaign Structure
As with Adwords, Search Ads will be organized into three levels: campaigns, ad groups, and keywords. A campaign is comprised of ad groups, and an ad group is made up of keywords.

This is good news since it will make structuring campaigns familiar and hopefully seamless. It sounds like Apple will be making an API and power tools available from the get-go, which should be music to performance marketers’ ears. It’s rare for new ad networks to offer these types of tools early on.

CPT and CPA Bidding
Search Ads will have the option to bid for a maximum cost per tap (CPT) or a target cost per install (CPA). CPA bidding will require historical conversion data in order to accurately deliver ads at the CPA target.

The auction system is of the second-price auction type, in which an advertiser must only pay whatever’s necessary to beat out their next best competitor, and not their full maximum bid. The bidding therefore also closely resembles the AdWords bidding options and bidding algorithm.

Keyword Research
Just like Google, Apple is offering keyword insights. Apple will produce lists of keyword suggestions and related keywords for base keywords input by advertisers. Additionally, Apple will display relative search volume for keywords, which will enable marketers to uncover the highest volume keywords.

Search Term Reporting

Without knowing the actual search terms used to get clicks or installs, it isn’t possible to effectively manage search advertising campaigns. Good thing that the Search Ads platform will make this data available, which will help advertisers see exactly what’s actually working and will be an invaluable source for expanding keywords. Equally important, search term reports include data on what isn’t working. Marketers can use this data to exclude irrelevant search terms with negative keywords.

Match Types: Broad and Exact Match
The standard broad match and exact match types will be available in Apple Search Ads. This allows for an ideal combination of broad match campaigns, which help to discover new converting search terms, and exact match campaigns, which make it possible to capitalize on consistently converting search terms. (Pro tip: just as with AdWords, don’t forget to use negative exact match keywords in broad campaigns to ensure the right search query triggers the right keyword.)

Geographical Targeting and Ad Scheduling
Advertisers will have the familiar option to target specific geographies as well as specific days and times with Search Ads. It’s not yet clear whether these targeting segments will be included in reporting. If the relative performance of different geographies, days, and times don’t end up being included in reporting (like you would find in the AdWords Dimensions Report), it could make the targeting options somewhat less effective from an optimization standpoint.

What’s Different


Only One Ad Result
Apple will only display one ad, at most, for each search query. This has pretty big implications from a competitive standpoint, and could lead to very high CPTs if there are many apps competing for the same keywords.

Apple has always been focused on user experience, which is likely the reason they’ll only allow one ad per query. Only allowing one ad is a conspicuous difference with Google, which shows multiple ads on its SERPs. On the other hand, Google did recently make the decision to stop showing ads on the right-hand side of the SERP, significantly reducing the number of ads that appear for each search query.

Relevance Floor Requirement
The single ad result poses a challenge since it could make it easy for app developers with a lot of resources to block everyone else out of the ad auction. To prevent this from happening, Apple will only display ads that are relevant to a user’s search query with a minimum relevance floor required to even be eligible for a given keyword search term auction. This also allows for a better user experience for the searcher, ensuring they only see relevant ads.

This means that if an app is not relevant to a search query, it won’t be shown no matter how high the bid. For example, a productivity app won’t be eligible to show an ad for the search term “zombie games” regardless of the bid. Apple will determine relevance according to the metadata on an app’s product page.

This is in clear contrast to Google, which allows advertisers to overcome low relevance (a.k.a. Quality Score) by compensating with higher bids. With AdWords, an advertiser can show up for any search query if the corresponding keyword has a high enough Ad Rank (the formula for Ad Rank is Quality Score multiplied by bid). It seems as if Apple will have a similar ad ranking process to determine which ad is shown, but with the significant difference of having a minimum requirement for relevancy.

No Control Over Ad Copy
Apple Search Ads will be generated automatically based on the metadata from an app’s product page. This is both a blessing and a curse. On the plus side, it will make for a very simple setup process. But it also means that marketers will have less room to customize ad copy, which means they’ll have less freedom to optimize since they will not be able to develop keyword-based ad copy for relevance or conduct A/B testing.

It’s uncertain at this point how accurately Apple will be able to generate a relevant ad since the ad will be wholly dependent on metadata from app product pages. App developers take note – it might be a good time to revisit your product page descriptions.

Other Targeting Options
Advertisers on Apple Search Ads will also be able to both target and view reporting on performance by age, gender, and device, options which are difficult to achieve with Google AdWords. This will allow for much better targeting, and will help improve efficiency for advertisers who take the time to analyze what’s performing best and optimize accordingly.

Install and Conversion-Funnel Tracking
While there will be install tracking, it’s not clear how far down the funnel Apple’s Attribution API will go, or whether Search Ads will play nicely with other attribution and event tracking systems. Good marketers care about what happens beyond the install, and an inability to track what happens down-funnel could limit the effectiveness of Search Ads.

Most mobile app ad networks take advantage of the vast ecosystem of third-party attribution and measurement tools, but it that might not be what Apple has in store, and this could be a challenge for app developers trying to optimize towards meaningful metrics like ARPU, LTV, and payback period.

Google AdWords integrates fairly well with third-party attribution tools and offers seamless install tracking for Android apps. It’s also capable of tracking in-app events for Android apps using the AdWords Conversion Tracking SDK. Hopefully, Apple will enable similar tracking capabilities, otherwise it will make optimization a very challenging task.

A Potential Gamechanger

Apple Search Ads will undeniably change the landscape for user acquisition on mobile. However, the effectiveness of the platform will depend on the ability for app developers to achieve enough reach (given the relevancy floor restriction on metadata discussed earlier), and also on the platform’s ability to track the efficacy of each keyword from taps all the way through to in-app engagement.

We’re excited to get started with Search Ads, and we plan to post our learnings and best practices once we’ve had a chance to test the platform and get some hard data. The beta for Search Ads starts today (you can opt-in here), and the full release will be later this fall.