name: review-daily-metrics
description: Queries main product metrics and surfaces highlights, anomalies, and trends. Use as part of the daily start-of-day routine.

Review Daily Metrics

When to Use

This skill is called by the start-of-day routine (step 4). It can also be run independently when asked for a quick metrics check.

Inputs

• Data guide: your data guide (read before querying)
• Product strategy context: your product strategy documentation (for interpreting metrics)

Tools

Data Warehouse MCP

Metrics to Track

These are the default daily metrics. This can be refined this list over time.

Note: The specific SQL queries depend on table schemas documented in your data guide. Always read the data guide before writing queries to use correct table names, join keys, and known gotchas.

Workflow

Step 1 -- Read the data guide

Read your data guide/schema documentation for current table schemas, metric definitions, and known data issues.

Step 2 -- Query metrics

For each metric in the table above, write and execute a SQL query. Compare today's value against:
- Yesterday (daily change)
- Same day last week (weekly trend)
- 7-day rolling average (smoothed trend)

Use your SQL query tool for well-defined queries. Use your data exploration tool if you need to explore or the data guide doesn't cover a specific metric.

Step 3 -- Identify highlights

Flag anything notable:
- Spikes or drops > 15% compared to the 7-day average
- Trend changes (e.g. 3+ consecutive days of decline)
- Records (highest/lowest in the past 30 days)
- Anomalies (unexpected patterns, data gaps)

Step 4 -- Produce summary

Write a concise metrics summary with highlights.

Output

Generate HTML content using the routine template at skills/system/routine-html-template.md. Read that file for the full CSS and component patterns.

If running as part of the start-of-day routine: append this section to the morning briefing HTML (after the Slack sections but before the footer). The summarise-slack-and-suggest-tasks skill assembles the combined file.

If running standalone: save as a standalone HTML file to your output folder/daily-metrics-YYYY-MM-DD.html and open it.

HTML structure for this skill

<hr class="divider">

<div class="section">
  <div class="section-title">Daily Metrics · [date]</div>

  <div class="metric-grid">
    <div class="metric-card">
      <div class="metric-value">[N]</div>
      <div class="metric-label">DAU</div>
      <div class="metric-change [up/down/flat]">[+/- %] vs 7d avg</div>
    </div>
    <div class="metric-card">
      <div class="metric-value">[N]</div>
      <div class="metric-label">Product sent</div>
      <div class="metric-change [up/down/flat]">[+/- %] vs yesterday</div>
    </div>
    <div class="metric-card">
      <div class="metric-value">[N]</div>
      <div class="metric-label">New signups</div>
      <div class="metric-change [up/down/flat]">[+/- %] vs 7d avg</div>
    </div>
    <div class="metric-card">
      <div class="metric-value">[%]</div>
      <div class="metric-label">Activation rate</div>
      <div class="metric-change [up/down/flat]">[+/- pp] vs 7d avg</div>
    </div>
    <div class="metric-card">
      <div class="metric-value">[%]</div>
      <div class="metric-label">Conversion to paid</div>
      <div class="metric-change [up/down/flat]">[+/- pp] vs 7d avg</div>
    </div>
  </div>
</div>

<div class="section">
  <div class="section-title">Highlights</div>
  <div class="card success">
    <div class="card-title">[Notable positive finding]</div>
    <div class="card-body">[Detail with numbers]</div>
  </div>
  <div class="card warning">
    <div class="card-label warning">Worth watching</div>
    <div class="card-title">[Emerging trend or anomaly]</div>
    <div class="card-body">[Detail]</div>
  </div>
</div>

If any metric can't be queried (missing table, data gap), note it with: <p class="quiet">Could not query [metric]: [reason]</p>

Guidelines

• Lead with the most important highlight, not the full table
• Be specific with numbers: "up 12%" not "increased"
• Flag data quality issues honestly (small samples, missing days, known pipeline delays)
• Read the data guide every time. Table schemas may have changed since the skill was written
• Keep the summary scannable. the user should understand the picture in 15 seconds

Change Log