Campaign Report Generator

You are generating a campaign performance report. Pulls the last 30 days, compares to the previous 30-day period, highlights anomalies, and ranks campaigns by performance.

Before you start, read the ads-playbook skill to load target CPA and business context.

---

Step 1: Pull Current Period Data (Last 30 Days)

SELECT
  campaign.name,
  campaign.status,
  campaign.advertising_channel_type,
  metrics.cost_micros,
  metrics.conversions,
  metrics.cost_per_conversion,
  metrics.clicks,
  metrics.impressions,
  metrics.ctr,
  metrics.average_cpc,
  metrics.search_impression_share,
  metrics.conversions_value
FROM campaign
WHERE segments.date DURING LAST_30_DAYS
  AND campaign.status = 'ENABLED'
ORDER BY metrics.cost_micros DESC

---

Step 2: Pull Previous Period Data (30-60 Days Ago)

Use explicit date ranges for the comparison period:

SELECT
  campaign.name,
  metrics.cost_micros,
  metrics.conversions,
  metrics.cost_per_conversion,
  metrics.clicks,
  metrics.impressions,
  metrics.ctr,
  metrics.average_cpc,
  metrics.search_impression_share
FROM campaign
WHERE segments.date BETWEEN '{30_days_ago}' AND '{60_days_ago}'
  AND campaign.status = 'ENABLED'
ORDER BY metrics.cost_micros DESC

Calculate the date range: if today is 2026-03-25, current period is 2026-02-23 to 2026-03-25, previous period is 2026-01-24 to 2026-02-22.

Cost conversion: Divide cost_micros, cost_per_conversion, and average_cpc by 1,000,000.

---

Step 3: Pull Daily Trend Data

For trend analysis and anomaly detection:

SELECT
  segments.date,
  metrics.cost_micros,
  metrics.conversions,
  metrics.clicks,
  metrics.impressions
FROM campaign
WHERE segments.date DURING LAST_30_DAYS
  AND campaign.status = 'ENABLED'
ORDER BY segments.date ASC

---

Step 4: Generate the Report

Structure the report as markdown. Save to the user's Desktop as campaign-report-YYYY-MM-DD.md.

Report Structure

---

Campaign Performance Report Period: [start date] to [end date] Generated: [today's date]

---

Account Overview

Metric	Current Period	Previous Period	Change
Total Spend	$X,XXX	$X,XXX	+X%
Conversions	XXX	XXX	+X%
CPA	$XX.XX	$XX.XX	-X%
Clicks	X,XXX	X,XXX	+X%
Impressions	XX,XXX	XX,XXX	+X%
CTR	X.X%	X.X%	+X.Xpp
Avg. CPC	$X.XX	$X.XX	+X%

Color-code changes in the chat:

• Improvements (lower CPA, higher conversions, higher CTR): positive
• Declines: flag them

Note whether CPA is above or below the target from the ads-playbook.

---

Campaign Breakdown

Campaign	Spend	Conv.	CPA	CTR	Imp. Share	vs Prev CPA
[name]	$X,XXX	XX	$XX.XX	X.X%	XX%	-X%
...	...	...	...	...	...	...
Total	$X,XXX	XXX	$XX.XX	X.X%

---

Top Performers (by CPA relative to target)

List the 3 campaigns with the best CPA (closest to or below target):

1. [Campaign] - CPA: $XX vs $XX target. X conversions. Why it works: [brief note on what's driving performance]

---

Bottom Performers (by CPA relative to target)

List the 3 campaigns with the worst CPA:

1. [Campaign] - CPA: $XX vs $XX target (Xx above target). $XXX spent. Potential issue: [brief diagnosis]

---

Anomalies and Trends

Scan the daily trend data for:

• Spend spikes or drops: Any day with spend >2x or <0.5x the daily average
• Conversion droughts: 3+ consecutive days with zero conversions in any campaign
• CTR shifts: Sudden CTR changes (>30% swing day-over-day)
• Impression share drops: Significant loss of impression share vs previous period

For each anomaly:

[Date]: [What happened] [Campaign name] saw a X% drop in conversions while spend remained flat. Possible causes: landing page issue, competitor activity, or audience fatigue.

---

Recommendations

Based on the data, provide 3-5 specific, actionable recommendations. Examples:

1. "Campaign X has CPA 2.5x above target. Run /mine-search-terms on it to check for wasted search terms."
2. "Campaign Y is budget-constrained (losing 35% impression share to budget). Run /budget-optimizer to evaluate an increase."
3. "Campaign Z has strong metrics but low volume. Consider expanding keywords or increasing bids."

Reference other skills in the plugin where relevant.

---

Step 5: Present and Save

1. Show the full report in-chat as formatted markdown
2. Save the report to the user's Desktop as campaign-report-YYYY-MM-DD.md
3. Confirm the file location

---

Edge Cases

• No previous period data: Skip the comparison columns. Note that this is the first reporting period.
• No MCP connected: Ask the user to paste campaign data or export from Google Ads. Generate the report from that.
• Single campaign: Still generate the full report structure, just with one row in the campaign breakdown.
• Meta Ads: Pull from Meta Ads API. Metrics map differently: use "results" instead of "conversions", "cost per result" instead of "CPA", "reach" and "frequency" instead of impression share.
• Apple Search Ads: Pull from the Campaign Report endpoint. Use "installs" instead of "conversions", "cost per install (CPI)" instead of "CPA", "share of voice" instead of "impression share".
• Multi-platform: If multiple platforms are connected, generate separate sections for each and include a combined summary at the top.