sm-analytics-review

SM Analytics Review — Monthly Performance Report

Overview

This skill pulls performance data from Metricool, analyzes what's working, identifies patterns in top-performing content, and produces recommendations for the next month's strategy. It runs monthly and saves insights to Cloud Brain to inform sm-content-calendar and sm-strategy-builder.

---

Pre-Flight — Load Context

1. mcp__cloud-brain__search_notes with query "social media strategy" — load strategy to compare against
2. mcp__cloud-brain__search_notes with query "approval history" — cross-reference with performance
3. mcp__cloud-brain__search_notes with query "social media discovery" — confirm active platforms and goals

---

Job Inputs (Ask Fresh Each Run)

Ask:

• Which month are we reviewing? (e.g., "June 2026")
• Any specific platform you want to focus on? (or "all of them")
• Any specific question you want answered from the data? (e.g., "why did my LinkedIn engagement drop in week 3?")

---

Step 1 — Pull Metricool Data

Using the Metricool MCP, retrieve data for the review period.

For each active platform, call the appropriate Metricool tool:

• Instagram: get_instagram_posts, get_instagram_reels, get_instagram_stories
• TikTok: get_tiktok_videos
• Facebook: get_facebook_posts, get_facebook_reels
• LinkedIn: get_linkedin_posts
• YouTube: get_youtube_videos
• Pinterest: get_pinterest_pins
• X/Twitter: get_x_posts

Also call:

• get_analytics — overall account metrics
• get_network_competitors — competitor context for benchmarking
• get_best_time_to_post — check if optimal times have shifted

Filter all calls to the review month's date range.

If Metricool is not connected: Ask the client to share their analytics manually — request screenshots or a summary of their top posts, reach, engagement, and follower growth for the month. Proceed with what they provide and note: "Connect Metricool (https://ai.metricool.com/mcp) to automate this data pull next month."

---

Step 2 — Top Content Analysis

From the data retrieved, identify:

Top 3 posts by engagement rate (likes + comments + shares / reach × 100):

• What platform
• What pillar
• What content type (video, graphic, text, carousel)
• What the hook or opening was
• Why it likely performed well (note patterns)

Bottom 3 posts by engagement rate:

• Same fields
• Why it likely underperformed

Best performing content type (e.g., Reels outperformed graphics 2:1)

Best performing content pillar (e.g., "Client Success" posts averaged 4.2% engagement vs. 1.8% for "Promotional")

---

Step 3 — Platform Performance Summary

For each active platform, calculate:

• Total posts published
• Average reach per post
• Average engagement rate
• Follower growth (net new followers this month)
• Best posting day and time (based on actual performance this month)
• Compared to strategy: On track / underperforming / outperforming

---

Step 4 — Goal Progress Check

Reference the client's goals from the discovery profile. For each primary goal:

Brand Awareness: Did reach / impressions grow month-over-month? Lead Generation: Did profile visits and link-in-bio clicks increase? Sales: Did any posts directly reference an offer? What was the response? Community Building: Did comment volume and follower growth trend up? Thought Leadership: Were any posts saved or shared at high rates?

Use available data to estimate progress. Note where data is unavailable.

---

Step 5 — Competitor Benchmarking

If Metricool competitor data is available (get_network_competitors_posts):

• Compare the client's average engagement rate against the top 3 competitors tracked
• Note any content types competitors used this month that the client hasn't tried
• Identify any competitor content that went unusually viral — note the topic and format

---

Step 6 — Recommendations

Based on all data, produce 3–5 specific, actionable recommendations for next month:

Format each recommendation as:

• What to do
• Why (data that supports it)
• How to apply (specific action in next month's calendar or strategy)

Examples:

• "Double down on Reels — they averaged 3.8% engagement vs. 1.2% for static posts. Replace 2 static posts/week with short Reels using the same educational content."
• "Post LinkedIn on Tuesday and Thursday only — Wednesday and Friday posts averaged 40% less reach."
• "Your 'Client Success' pillar is outperforming every other pillar. Add a 4th weekly post specifically for client wins or testimonials."

---

Save Insights to Cloud Brain

mcp__cloud-brain__write_note
path: brain/social-media/[business-name-slug]-analytics-[YYYY-MM].md

Include: raw summary stats, top/bottom posts, recommendations, and date of review.

---

Output — Monthly Analytics Report

---

Social Media Performance Report — [Business Name] — [Month Year] Pulled from Metricool · Reviewed: [date]

---

📊 At a Glance

Platform	Posts	Avg Reach	Avg Engagement	Follower Growth
Instagram	[n]	[n]	[%]	+[n]
LinkedIn	[n]	[n]	[%]	+[n]
TikTok	[n]	[n]	[%]	+[n]
Facebook	[n]	[n]	[%]	+[n]

---

🏆 Top Posts This Month

1. [Post topic] — [platform] — [engagement rate]% Why it worked: [reason]

2. [Post topic] — [platform] — [engagement rate]% Why it worked: [reason]

3. [Post topic] — [platform] — [engagement rate]% Why it worked: [reason]

---

📉 What Underperformed

[Brief notes on low performers — no judgment, just patterns]

---

📈 What's Working

• Best content type: [type]
• Best content pillar: [pillar]
• Best platform: [platform]
• Best posting day/time: [day/time]

---

🎯 Goal Progress

Goal	Status	Notes
[Goal]	[On track / Lagging / Ahead]	[data point]

---

📋 Recommendations for Next Month

1. [Recommendation with data reasoning]
2. [Recommendation with data reasoning]
3. [Recommendation with data reasoning]

---

Run sm-content-calendar next to apply these recommendations to next month's plan.