Apple Search Ads Account Structure: Semantics-Based vs Value-Based
A weak foundation limits even a well-run Apple Search Ads account. How campaigns, ad groups and keywords are arranged affects ad relevance, how clearly performance can be read, and whether spend can be scaled profitably. This guide compares two established layouts — semantics-based and value-based — and the six factors to settle before building either.
Decide these six factors first
Before choosing a layout, work through these six decisions. They determine which structure will actually fit your spend, markets and goals.
Campaign goals: Four common clusters are Brand protection (defend your own brand terms), Competitor (bid on rival brand terms), Prospecting (raise awareness with relevant users) and Discovery (find keywords you do not yet use). A simple account can be built directly on these four goals, named Brand, Competitors, Generic and Discovery. That works well for low spend, but complex goals such as pre-order testing need a different framework.
Countries and regions: Give each country its own campaign. Adapt keywords to the local language, account for cultural differences in what people search for, and localise app metadata for each market.
Ad relevance: Technical relevance covers how close the sponsored ad sits to the keyword and is driven by app category and metadata outside the account; custom product pages can strengthen it for specific audience segments. Actual keyword relevance covers how well keyword meaning matches the ad banner and is managed inside the account through ad groups combining keywords, creatives and audience.
Past performance: Do not mix keywords with different performance in one ad group, or the blended statistics will obscure analysis. Group keywords by past performance, for example by the lifetime value of the users behind them.
Budget: Budget is set at campaign level, so fund top-performing keywords first and spread what remains across less important ones. Separate campaigns by performance level make that waterfall easier to see and control. If monthly spend is under $1000, avoid an over-complex layout; reserve more sophisticated structures for larger budgets where flexibility pays off.
Audience segmentation: You can target new users and returning users. Build around your core audience, but do not ignore the general audience — especially for narrowly focused apps, where pursuing it can add impressions and support App Store ranking.
Semantics-based structure: five campaigns and a weekly transfer loop
This layout organises the account by what keywords mean, and is designed to grow impressions from high-performing keywords. It uses five campaign types.
Brand campaigns rest on your own branded keywords, reaching users with the highest intent and protecting the brand. Generic campaigns use descriptive, non-brand keywords aimed at people looking for a solution rather than a specific brand. Competitor campaigns bid on rivals' branded keywords to capture users searching by a competitor's name. Discovery campaigns mine new keywords using broad match and Search Match. Proxy campaigns test keywords surfaced by Discovery and show how they perform in mobile measurement partners.
Setup flow: Add the keywords already used in Brand, Generic and Competitor campaigns to Discovery as exact-match negatives so the account does not suffer data dilution. Review search terms in Discovery that lack enough performance data and add them to Proxy as exact matches. Then check performance: move strong keywords into Brand, Generic or Competitor campaigns, and pause weak ones. Repeat this transfer weekly.
In Brand, Generic and Competitor campaigns, create a separate ad group for each audience type — new users, returning users and all users. Leave Discovery and Proxy as they are. If you use theme ad groups based on search intent, split each theme by audience as well. For budget, emphasise branded keywords first, then waterfall the remaining spend across the other keywords.
Value-based structure: tiers by lifetime value
This layout groups keywords by the value of the users they bring, where value is tied to user intent and lifetime value. Its core components are a Brand campaign, a Discovery campaign and Tier campaigns, with an optional Proxy campaign.
Brand gets the maximum bid to acquire users most likely to become high-value customers. Tier 1 holds exact-match keywords bringing high lifetime-value users, with a competitive bid to maximise traffic. Tier 2 holds medium-value keywords with a medium bid. Tier 3 holds the lowest-value keywords with the lowest bid, and further tiers can be added if useful. Organising keywords by lifetime-value band with matching bids makes the account easier to navigate and control.
Discovery should be bid similarly to Tier 1 or Tier 2 so it stays competitive enough to surface new search terms. When new keywords appear, move them into the tier matching their value as exact matches, and mark them as negatives in Discovery. If you add a Proxy campaign, add its search terms as exact matches. For budget, use the waterfall approach: spread spend down the pyramid, prioritising the keywords that bring the most value.
The trade-off is maintenance. High-value keywords may have low scale while lower-value ones scale more easily, keyword value changes over time and must be re-checked, and the structure can feel overwhelming on a limited budget. Handled well, it can deliver higher return on investment and more effective scaling.
Choosing between them
Semantics-based is the easier build and the more time-tested of the two. Value-based is harder to create and maintain because keyword value must be watched continuously, but it offers stronger control over bids and budget by value. Weigh your budget size, how reliably you can estimate lifetime value by keyword, and how much ongoing optimisation you can sustain — then pick the layout that fits those constraints rather than the more complex one by default.
| Campaign goal | What it does | Typical campaign name |
|---|---|---|
| Brand Protection | Defend your brand against competition | Brand |
| Competitor | Bid on competitor keywords | Competitors |
| Prospecting | Increase brand awareness and engage relevant users | Generic |
| Discovery | Mine new keywords | Discovery |
Simple goal-based layout; best suited to simple accounts with low ad spend.
| Structure | Organising principle | Budget emphasis | Main risk |
|---|---|---|---|
| Semantics-based | Keyword meaning: Brand, Generic, Competitor, Discovery, Proxy | Fund branded keywords first, waterfall the rest | Needs weekly keyword transfers to stay optimised |
| Value-based | User value: Brand, Discovery, Tiers by lifetime value, optional Proxy | Waterfall spend down the value pyramid | Keyword value shifts over time; harder to maintain |
Comparison of the two approaches described in the guide.
Frequently asked questions
Should I choose semantics-based or value-based structure?
Use the semantics-based layout when you want a simpler, proven way to organise by keyword meaning and grow impressions. Use the value-based layout when you can estimate lifetime value by keyword and want budget and bids to follow that value, accepting more ongoing maintenance.
Can one campaign cover multiple countries?
Yes. Each country needs its own campaign, with keywords adjusted for local language, cultural differences in search behaviour, and localised app metadata.
Why separate keywords by past performance?
Keep strong and weak performers apart. Mixing them in one ad group blends the statistics and makes it harder to see what to optimise. Grouping by past performance or lifetime value keeps the analysis clean.
What is the difference between technical and actual keyword relevance?
Technical relevance is how closely the sponsored ad matches the keyword, shaped by app category and metadata outside the account. Actual keyword relevance is how well keyword meaning fits the ad banner, which you influence inside the account through the combination of keywords, creatives and audience in ad groups.