15 Meta Ads Prompts for Claude (Advantage+ CAPI Audiences)

Claude reads your full account structure through PaidSync and surfaces structural problems across campaigns, audiences, and tracking. It ranks issues by impact so you can address the most expensive problems first. Good starting point for any account you have not audited in 30+ days.

Claude pulls current delivery and ROAS data for every active campaign and identifies the ones that have gone dark or are underperforming against a 2x minimum threshold. Campaigns that stop spending mid-month often have budget exhaustion or delivery issues that are not surfaced in the Ads Manager summary view.

High frequency combined with a declining CTR is the clearest signal that your audience has seen your ads too many times. Claude identifies the ad sets showing this pattern and gives you a concrete recommendation for each one rather than just flagging the numbers.

Audience overlap between your own ad sets raises your CPMs because you are bidding against yourself. Claude identifies the highest-overlap pairs and recommends consolidation or audience exclusion rules that reduce self-competition without reducing overall reach.

Advantage+ Shopping uses Meta's automated targeting and placement system and tends to outperform manual Shopping campaigns for established e-commerce accounts with good pixel data. Claude configures the campaign with your catalog connection, budget, and ROAS target, then shows you every setting before creating the campaign in your account.

Claude sets up the full three-ad-set structure with the targeting, exclusions, budgets, and creative duplication you specified. It shows the complete configuration before building anything so you can confirm audiences and budgets are correct. This kind of structured test is typically a 30-minute task in Ads Manager done in one conversation.

Claude queries performance data across all your conversion campaigns, applies the spend and ROAS filters you set, and shows you every qualifying ad set before touching anything. This is the fastest way to cut budget waste across a large account where checking each ad set manually would take an hour.

Lookalike audiences built from purchase events are typically the highest-performing cold audiences in direct-response campaigns. Claude creates all three sizes and reports estimated reach before building them so you can decide whether the 5% size is large enough to be worth the lower similarity score.

A clean funnel-based retargeting audience structure is one of the highest-ROI setups in any Meta account. Claude creates all four audiences with the correct pixel event rules and exclusions, then names them consistently so your Ads Manager stays organized. Each audience is sized and confirmed before being built.

Claude uses the audience insights data from your custom audience to identify the demographic and interest patterns that are strongest among your existing visitors. This gives you a data-backed starting point for cold interest targeting rather than guessing at relevant categories.

Event Match Quality (EMQ) directly affects how well Meta can attribute your CAPI events to users. A score below 6 means you are sending insufficient customer data and conversions are being lost in attribution. Claude reads your current CAPI configuration and tells you exactly which parameters to add to improve each event's score.

Deduplication errors inflate your reported conversion numbers and corrupt Smart Bidding signals. Claude checks whether your event_id values match between browser and server events and whether Meta is correctly deduplicating them. Missing or inconsistent event_id is the most common cause of inflated purchase counts.

Claude fetches both weeks of data and organizes it by objective type so you can see whether your prospecting, retargeting, and conversion campaigns are each performing against their own benchmarks. The week-over-week comparison surfaces whether changes in one objective are affecting the others.

Claude aggregates creative-level performance data and produces a ranked list with scaling and retirement recommendations based on ROAS and frequency. High-ROAS, low-frequency creatives are candidates for budget expansion. High-frequency, declining-ROAS creatives are candidates for replacement.

Cost cap and bid cap behave differently in Meta's auction. Cost cap aims for average cost efficiency and may underspend. Bid cap limits individual bid amounts and can fail to spend if set too low. Claude pulls the side-by-side comparison so you can make an informed choice about which strategy to consolidate onto based on your actual data.