Diagnose Meta Campaign Performance Gaps

Business Context

A large advertiser is running conversion campaigns across Meta Ads Manager using Facebook Feed, Instagram Reels, and Stories placements. Over the last 4 weeks, spend increased, but reported business outcomes did not keep pace, and the Account Executive wants a structured way to analyze campaign performance and identify where the gaps are.

Metric Scenario

Last month, the advertiser spent $1.2M and generated 24M impressions, 360K link clicks, 18K landing page views, 7,200 purchases, and $864K in attributed purchase value. This month, spend rose to $1.5M, impressions to 30M, and clicks to 420K, but landing page views only increased to 19K, purchases fell to 6,300, and attributed purchase value declined to $756K. CTR moved from 1.50% to 1.40%, CPC from $3.33 to $3.57, landing page view rate from 5.0% to 4.5% of clicks, purchase conversion rate from 40.0% to 33.2% of landing page views, and ROAS from 0.72x to 0.50x.

Stakeholders are asking whether the issue is audience quality, creative fatigue, placement mix, on-site conversion friction, or attribution/reporting changes.

Requirements

1. Define the primary KPI you would use to evaluate campaign performance and explain why.
2. Decompose the performance drop into funnel stages and quantify the largest gap.
3. Identify the most important cuts of the data to isolate whether the issue is in delivery, engagement, or conversion.
4. Propose 3-5 hypotheses for the decline and how you would validate each.
5. Recommend immediate actions for the advertiser and the Meta account team.

Data Available

• Meta Ads Manager delivery logs: campaign_id, ad_set_id, ad_id, objective, spend, impressions, reach, frequency, clicks, landing_page_views, purchases, purchase_value
• Placement performance table: Facebook Feed, Instagram Feed, Instagram Reels, Stories, Audience Network
• Creative diagnostics table: creative_id, format, thumb-stop rate, video plays, hold rate, outbound CTR
• Meta Pixel / Conversions API event logs: page_view, view_content, add_to_cart, initiate_checkout, purchase, event_time, browser, device, event_match_quality