Analytics - Nectr

The Analytics page provides deep insights into team performance, code quality trends, and developer contributions with interactive charts and data visualizations.

Overview

Analytics is organized into four tabs:

Overview - High-level metrics and activity charts
Team - Contributor profiles and strengths
Repos - Repository-level statistics
Insights - Confidence scores and merge time analysis

Filter Bar

All tabs share a unified filter system:

Repository Filter

Only connected repositories appear in this dropdown. Disconnected repos are excluded.

All repositories (default)
List of connected repos (e.g., owner/repo-name)

Author Filter

All authors (default)
List of authors from analytics data (e.g., @username)

Time Range Filter

This week (7 days)
Last 30 days (default)
Last 90 days

Active Filters

When filters are applied, chips appear showing:

[owner/repo ✕]  [@username ✕]

Clear all

Click the ✕ to remove individual filters or “Clear all” to reset.

Overview Tab

Key Performance Indicators

Four KPI cards display:

Total PRs

Total number of pull requests in the selected timeframe

Success Rate

Percentage of PRs successfully merged

Avg Merge Time

Average hours from PR open to merge

Avg PR Size

Average lines of code changed per PR

PR Size Area Chart

A multi-line area chart showing:

Green area: Lines added (additions)
Red area: Lines deleted (deletions)
Teal area: Net change (absolute difference)

Chart features:

X-axis: Date (MM/DD format)
Y-axis: Lines of code (abbreviated: 1k, 2k, etc.)
Interactive tooltip showing exact values
Gradient fills for visual clarity

Bottom stats:

Average Additions | Average Deletions | Total PRs
    45,678       |      12,345       |    47

Activity Timeline

Line chart showing review activity over time:

Amber line: Total reviews
Green line: Completed reviews
Red line: Failed reviews

Features:

Smooth curve (monotone interpolation)
X-axis shows date range based on filter (7/30/90 days)
Y-axis shows review count
Hover tooltip with exact values

Verdict Pie Chart

Circular donut chart showing AI decision distribution:

Green: Approved PRs
Red: Changes Requested
Amber: Needs Discussion

Legend below chart:

● Approved            [████░░░░░░] 23
● Changes Requested   [██░░░░░░░░] 8
● Needs Discussion    [██░░░░░░░░] 5

Each row shows:

Colored dot indicator
Verdict name
Horizontal progress bar
Total count

Team Tab

Team data requires at least one connected repository. If no repos are connected, an empty state appears.

Contributor profile cards display:

Card Layout

Avatar: Colored circle with first letter of username
Username: @username with bold styling
PR Badge: Count of PRs by this author
Profile Summary: AI-generated description (if available)
Strengths: Green badges showing areas of expertise
Patterns: Amber badges showing recurring behaviors

Example strengths:

✓ Strong API design
✓ Excellent test coverage
✓ Clean TypeScript types

Example patterns:

⚠ Tends to create large PRs
⚠ Frequent async/await issues

Empty State

If no contributor data exists:

👥 No contributor data yet

Connect a repo and complete some reviews.

Repos Tab

Table view of repository statistics:

Repository	PRs	Merged	Issues
owner/repo	47	42 (89.4%)	12
other/repo	23	18 (78.3%)	5

Column details:

Repository: Full repo name in monospace
PRs: Total PR count
Merged: Count and percentage in parentheses
Issues: Total issues flagged (red if > 0)

Hover effect: Row background changes to #191919

Empty State

If filtering by a specific repo with no data:

No data for owner/repo.

Insights Tab

Confidence Distribution

Bar chart showing AI confidence scores (1-5):

Red bar (1/5): Very low confidence
Orange bar (2/5): Low confidence
Amber bar (3/5): Medium confidence
Light green bar (4/5): High confidence
Green bar (5/5): Very high confidence

Interpretation:

High confidence (4-5)

The AI is very certain about its recommendations. Likely based on clear patterns and best practices.

Medium confidence (3)

The AI has some uncertainty. May require human judgment or team discussion.

Low confidence (1-2)

The AI detected complex or ambiguous code. Strongly recommend manual review.

Merge Time Stats

Four stat boxes showing time-to-merge metrics:

Avg

Mean time from PR open to merge

Median

50th percentile merge time (less affected by outliers)

Fastest

Shortest merge time in dataset

Slowest

Longest merge time in dataset

All values displayed in hours (e.g., “12.5h”).

Author Performance Table

Detailed per-developer breakdown:

Author	PRs	Merged	Issues	Confidence
@alice	28	26	3	4.2/5
@bob	19	15	8	3.8/5

Column descriptions:

Author: GitHub username
PRs: Total PRs opened
Merged: Successfully merged PRs (green)
Issues: Issues flagged by AI (red)
Confidence: Average AI confidence score

Sorting: Currently manual, sorted by PR count descending

Data Refresh

All analytics data refreshes automatically when:

Filter values change
A new PR review completes
Page regains focus after 5+ minutes

Use the browser’s refresh button to force an immediate data reload.

Loading States

While fetching analytics:

KPI cards: 4 skeleton rectangles (28px height)
Charts: Full-height skeletons with rounded corners
Tables: Multiple skeleton rows

All skeletons have a subtle pulse animation.

Technical Details

API Endpoints

GET /api/analytics/summary
GET /api/analytics/timeline?days={days}
GET /api/analytics/insights?days={days}
GET /api/analytics/contributors?repo={repo}
GET /api/repos

Data Models

interface AnalyticsSummary {
  total_reviews: number;
  success_rate: number;
  avg_processing_seconds: number;
  avg_merge_hours: number | null;
  avg_pr_size: number | null;
  last_week_activity: WeeklyPR[];
}

interface AnalyticsInsights {
  period_days: number;
  verdicts: { APPROVE: number; REQUEST_CHANGES: number; NEEDS_DISCUSSION: number };
  confidence_distribution: Record<string, number>;
  merge_time: MergeTimeStats;
  per_author: AuthorStats[];
  per_repo: RepoStats[];
}

interface Contributor {
  username: string;
  profile_summary: string | null;
  patterns: string[];
  strengths: string[];
  pr_count: number;
  last_seen_pr: number | null;
}

Chart Library

All charts use Recharts with custom styling:

Dark theme (#1A1A1A backgrounds)
Monospace font (Geist Mono) for labels
Responsive containers (100% width)
Custom tooltips with brand colors

Best Practices

Start Broad

Begin with “All repositories” and “All authors” to see overall trends.

Drill Down

Apply filters to investigate specific repos or developers.

Compare Timeframes

Toggle between 7/30/90 days to identify seasonal patterns.

Watch Confidence

Low confidence scores may indicate areas needing better documentation.

Common Insights

High Merge Times

If average merge time exceeds 24 hours:

Check if PRs are too large (use PR Size chart)
Look for bottlenecks in review process
Consider setting team response time goals

Low Success Rates

If success rate drops below 80%:

Review failed reviews in the Reviews page
Check for common error patterns
Ensure proper branch protection rules

Unbalanced Workload

If one author has 3x more PRs than others:

Consider redistributing work
Check if certain areas lack expertise
Use Team tab to identify knowledge gaps

Get Started

Core Features

Dashboard

​Overview

​Filter Bar

​Repository Filter

​Author Filter

​Time Range Filter

​Active Filters

​Overview Tab

​Key Performance Indicators

Total PRs

Success Rate

Avg Merge Time

Avg PR Size

​PR Size Area Chart

​Activity Timeline

​Verdict Pie Chart

​Team Tab

​Card Layout

​Empty State

​Repos Tab

​Empty State

​Insights Tab

​Confidence Distribution

​Merge Time Stats

Avg

Median

Fastest

Slowest

​Author Performance Table

​Data Refresh

​Loading States

​Technical Details

​API Endpoints

​Data Models

​Chart Library

​Best Practices

Start Broad

Drill Down

Compare Timeframes

Watch Confidence

​Common Insights

​High Merge Times

​Low Success Rates

​Unbalanced Workload

Build docs developers (and LLMs) love

Overview

Filter Bar

Repository Filter

Author Filter

Time Range Filter

Active Filters

Overview Tab

Key Performance Indicators

PR Size Area Chart

Activity Timeline

Verdict Pie Chart

Team Tab

Card Layout

Empty State

Repos Tab

Empty State

Insights Tab

Confidence Distribution

Merge Time Stats

Author Performance Table

Data Refresh

Loading States

Technical Details

API Endpoints

Data Models

Chart Library

Best Practices

Common Insights

High Merge Times

Low Success Rates

Unbalanced Workload