How the health score is calculated and how the AI generates prioritized recommendations from your Meta ad account data.
How Campaiyn's AI works
Campaiyn is not a rules engine that flags obvious errors. It analyzes your entire Meta ad account across five health dimensions and uses AI to prioritize the actions that will have the most impact on your specific account.
What is the health score?
Your health score is a number from 0 to 100 that measures how well your Meta ad account is performing. It is calculated across 5 pillars:
Budget Efficiency -- How well your budget is allocated. Are campaigns over-spending on underperformers? Is spend concentrated in Learning Limited ad sets?
Creative Performance -- The health of your ad creatives. This includes frequency, CTR trends, creative age, and variety across ad sets.
Campaign Structure -- Whether your account follows the structural patterns that Meta's algorithm handles best. Too many campaigns, overlapping audiences, single-creative ad sets -- these all lower this score.
Audience Quality -- The quality of your targeting configuration. Overly narrow audiences, audience overlap between ad sets, and missing Advantage+ signals are assessed here.
Delivery Health -- Whether your ads are actually reaching people effectively. Learning Limited status, high frequency, policy flags, and delivery restrictions all affect this pillar.
Each pillar gets its own score. Your overall health score is a weighted average of all five.
How the AI generates recommendations
When you run a scan, Campaiyn:
Pulls your campaign, ad set, ad, and event data from Meta's API
Analyzes the data against known patterns and benchmarks across all 5 pillars
Identifies which issues are most likely to be costing you money or limiting your results
Uses AI to generate specific, contextual recommendations -- not generic advice, but actions written with your actual data in mind
Ranks the recommendations by estimated impact
Each recommendation includes:
The specific issue that was found (e.g., "Ad set X has been Learning Limited for 14 days")
The context explaining why it matters
A clear, actionable next step
An estimated impact on your score or performance
Note
The AI does not generate the same recommendations for every account. Two accounts with the same score can get completely different recommendations, because the causes are different. The AI interprets your data in context.
Does the AI learn from my account over time?
Yes. As you mark recommendations as done or apply them via Apply Now, Campaiyn records which actions led to score improvements on subsequent scans. This feedback loop helps the AI calibrate the importance of different recommendations for accounts similar to yours.
What data does Campaiyn access?
Campaiyn reads the following data from Meta's Marketing API:
Campaign names, objectives, budgets, and status
Ad set targeting, budgets, schedules, and delivery status
Ad creative names, formats, and performance metrics
Account-level event and pixel configuration
Campaiyn does not access your billing information, personal Business Manager settings, or any data outside of ad performance.