Power User Scoring

This drill builds a scoring model that identifies your top users — the ones whose behavior signals deep product investment and advocacy potential. The output is a ranked list of power users in your CRM with composite scores, ready to recruit into an advocacy program.

Prerequisites

• PostHog tracking core product events for at least 30 days
• Attio configured with contact and company records
• Clear understanding of which product actions correlate with retention and expansion

Steps

1. Define the scoring dimensions

Query PostHog to identify the behaviors that differentiate your most valuable users. Build the composite score from these dimensions, each weighted 0-100:

Usage depth (weight: 30%) Using posthog-cohorts, compare users in the top decile of session frequency and feature breadth against the median. Measure:

• Distinct features used in last 30 days
• Advanced feature usage (API, integrations, automations, admin functions)
• Sessions per week (trailing 4 weeks)
• Average session duration

Score formula: (distinct_features / max_features) * 0.4 + (advanced_feature_count / max_advanced) * 0.3 + (sessions_per_week / 7) * 0.2 + min(avg_session_minutes / 30, 1) * 0.1

Tenure and consistency (weight: 20%) Using posthog-custom-events, calculate:

• Account age in days
• Weekly active weeks out of total weeks (consistency ratio)
• Streak: consecutive weeks with at least 3 sessions

Score formula: min(account_age_days / 180, 1) * 0.3 + consistency_ratio * 0.5 + min(current_streak_weeks / 12, 1) * 0.2

Collaboration signal (weight: 20%) Measure team and social behaviors:

• Team members invited
• Shared assets created (shared dashboards, templates, reports)
• Comments or collaboration actions taken

Score formula: min(teammates_invited / 5, 1) * 0.4 + min(shared_assets / 10, 1) * 0.3 + min(collab_actions_30d / 20, 1) * 0.3

Expansion signal (weight: 15%) Measure growth trajectory:

• Usage volume month-over-month growth rate
• Feature tier (free, starter, pro, enterprise)
• Approaching plan limits (boolean: at 80%+ of any limit)

Score formula: min(mom_growth_rate / 0.5, 1) * 0.4 + tier_score * 0.3 + (approaching_limit ? 1 : 0) * 0.3

Sentiment signal (weight: 15%) Incorporate explicit feedback:

• NPS score (if available, normalize 0-10 to 0-1)
• Support ticket sentiment (positive interactions / total interactions)
• Feature request submissions (signals investment in the product roadmap)

Score formula: nps_normalized * 0.5 + support_sentiment_ratio * 0.3 + min(feature_requests / 3, 1) * 0.2

Composite score: depth * 0.30 + tenure * 0.20 + collaboration * 0.20 + expansion * 0.15 + sentiment * 0.15

2. Implement the scoring pipeline

Using posthog-custom-events, create a computed person property power_user_score that updates daily. Fire an event power_user_score_computed with the composite score and all dimension scores as properties:

posthog.capture('power_user_score_computed', {
  composite_score: 78,
  depth_score: 85,
  tenure_score: 72,
  collaboration_score: 65,
  expansion_score: 80,
  sentiment_score: 90,
  percentile: 94
});

3. Create power user cohorts

Using posthog-cohorts, define tiered cohorts:

• Champions (score >= 80, top ~5%): deepest users, highest advocacy potential
• Power Users (score 60-79, top ~15%): strong candidates for program recruitment
• Rising Stars (score 40-59): users trending upward, worth nurturing toward power-user status
• Standard Users (score < 40): not yet candidates

4. Sync scored users to CRM

Using attio-custom-attributes, create a power_user_score attribute on contacts and a power_user_tier attribute (Champion / Power User / Rising Star / Standard). Using attio-lists, create a "Power User Candidates" list filtered to score >= 60. This list feeds the advocacy program recruitment pipeline.

5. Build the scoring dashboard

Build a PostHog dashboard with:

• Distribution histogram of power user scores (are they normally distributed or bimodal?)
• Tier breakdown: count and percentage in each tier
• Score trend over time: is the population getting more or less engaged?
• Top 20 power users ranked by composite score with dimension breakdown
• Rising Stars with highest month-over-month score increase (these are your acceleration targets)

6. Validate and calibrate

Review the top 20 scored users manually. Ask:

• Do these look like your best users? Would you want them as advocates?
• Are any obvious power users missing? If so, a scoring dimension is underweighted.
• Are any low-value users scoring high? If so, a dimension is overweighted.

Adjust weights and re-run until the top 20 matches your intuition. Lock weights and document the calibration.

Output

• Composite power user score computed daily for all active users
• 4-tier cohort system in PostHog (Champions, Power Users, Rising Stars, Standard)
• Scored user list synced to Attio with tier labels
• Scoring dashboard in PostHog
• Calibrated and documented scoring weights

Triggers

Run the scoring pipeline daily via n8n cron or PostHog computed properties. Re-calibrate weights quarterly or when the product's core actions change significantly.