borghei/Claude-Skills / project-management/execution/activation-funnel
activation-funnel
VerifiedDesigns and analyzes activation funnels: stage conversion, drop-off, bottlenecks and the aha event that predicts retention. Use when a funnel leaks, signups grow but activation is flat, or you must pick an activation event.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 9.6 KB |
| assets/activation_metric_worksheet.md | markdown | 5.1 KB |
| assets/funnel_design_canvas.md | markdown | 4.5 KB |
| assets/sample_funnel.json | config | 1.5 KB |
| examples/freemium-saas-day-7-retention-diagnosis.md | markdown | 15.1 KB |
| references/activation-aha-moment-patterns.md | markdown | 10.0 KB |
| references/funnel-analysis-playbook.md | markdown | 11.1 KB |
| references/pirate-metrics-deep-dive.md | markdown | 10.2 KB |
| references/red-flags.md | markdown | 13.5 KB |
| scripts/funnel_analyzer.py | script | 17.4 KB |
references/activation-aha-moment-patterns.md
View on GitHub# Activation "Aha Moment" Patterns
A working catalogue of how real products defined their activation events. Use these as inspiration -- not as templates to copy. Each activation event is specific to a product's value model.
## What makes a good activation event
Sean Ellis's framework: the activation event is the action that, statistically, separates users who retain from users who churn. The activation event is:
1. **Specific.** A count and an action and a window. "Used the product" is not specific. "Sent 3 messages in the first week" is.
2. **Observable.** It can be instrumented and queried in your event store.
3. **Predictive.** Users who do it retain at significantly higher rates than users who do not.
4. **Achievable in a short window.** The whole point is to optimize the path to it; if it takes 60 days to reach, it is too far out.
5. **Self-discovered.** The user does it because the product is useful, not because of a prompt or a coupon.
Anti-pattern: "completes onboarding" is not an activation event. It is the output of an onboarding flow. The activation event is what the user does *after* onboarding that signals real value.
## Worked examples
### Slack: 2,000 messages sent by a team
Slack's much-cited activation event. Stewart Butterfield and team observed that teams who hit ~2,000 messages were unlikely to churn -- it indicated genuine workflow adoption, not just a trial.
**Why it works:**
- Specific (count + action + scope).
- Achievable for active teams in ~30 days.
- Predictive of long-term retention.
- Self-discovered: users send messages because they need to, not because Slack prompts them.
**PM lesson:** define activation at the **team** level for collaboration products. Per-user activation misses that the product creates value in groups.
### Facebook: 7 friends in 10 days
Famous activation event from Facebook's growth team (Chamath Palihapitiya era). Users who added 7 friends within 10 days of signup retained dramatically better.
**Why it works:**
- Specific count + time window.
- Captures the network-effect threshold; below 7 friends, the feed is empty and Facebook is useless.
- Optimizing for it drove the famous "People You May Know" + import-contacts features.
**PM lesson:** for network products, activation is often a network-size threshold. Find the inflection in the retention curve.
### Dropbox: 1 file in 1 folder on 1 device
Dropbox's activation event. Sounds trivial but is exactly right: a user with one synced file on one device has experienced the cloud-storage value. The metric drove the entire onboarding UX.
**Why it works:**
- Captures the "I get it now" moment for cloud storage.
- Achievable in a single session.
- Drives a UX optimization (the onboarding flow ends when the user has a file synced).
**PM lesson:** activation is sometimes a single-session event, not a multi-day count. Match the event to the value model.
### Airbnb (split: host and guest)
| Side | Activation |
|---|---|
| Host | First successful booking (a real guest stays, a real review is left) |
| Guest | First completed trip (a stay, a review left) |
**Why it works:**
- Two-sided marketplaces need two activation definitions.
- Each side's activation requires the other side to engage -- this drives liquidity strategy.
**PM lesson:** marketplaces have at least two activation events, one per side. Tracking only one half hides the supply or demand bottleneck.
### Notion: 3+ pages created + 1+ collaborator invited
Notion's activation centers on creating real content and inviting at least one collaborator. The single-user "I made a page" is too weak; the collaborative "we made pages together" is sticky.
**Why it works:**
- Combines content depth (3 pages = the user is actually using it for real work) with network effect (1 collaborator = someone else can see it).
- Notion's onboarding now optimizes heavily for this combined event.
**PM lesson:** activation can be a composite (count of X **and** count of Y). Composites are noisier to instrument but often more predictive.
### Spotify: 25+ minutes of listening in week 1
Spotify's activation centers on minutes-listened in the first week. Users above this threshold retain dramatically better.
**Why it works:**
- Music consumption is the value; minutes is the right unit.
- A first-week window matches the trial-decision horizon.
- Drives the onboarding focus on quick playlist creation and recommendation quality.
**PM lesson:** for attention products, activation is usually a "time-spent" threshold within a short window.
### Duolingo: 1 lesson on 3 of the first 7 days
Duolingo's activation looks at consistency of use in the first week, not just total volume. A user who completes 7 lessons on day 1 is not as predictive as one who completes 3 lessons across 3 different days.
**Why it works:**
- Captures habit formation, which is Duolingo's core value model.
- Drives the famous streak-mechanic and daily-reminder UX.
**PM lesson:** for habit / learning products, activation is a frequency pattern, not a volume threshold.
### Twitter / X: 30+ accounts followed
Twitter's classic activation event was following 30+ accounts. Users below this threshold had a near-empty timeline and rarely returned.
**Why it works:**
- Network-effect threshold: below 30, the timeline does not feel alive.
- Drove the onboarding "suggest 10 accounts" UX.
**PM lesson:** for feed / consumption products, activation often requires a critical mass of subscribed sources.
### HubSpot: 5 contacts added + 2 deals tracked
HubSpot's activation for the CRM product. Users who set up at least 5 contacts and 2 deals are using it for real sales work, not just kicking the tires.
**Why it works:**
- B2B SaaS activation typically requires multi-object setup -- one record is not enough to signal commitment.
- The threshold matches the smallest realistic sales pipeline.
**PM lesson:** B2B activation events often require setup-completion across multiple object types. Single-object thresholds are too easy.
### Pinterest: 5 boards followed + 1 pin saved
Pinterest's activation combines a feed-source threshold (5 boards) with a single contribution (1 pin saved). Together they signal real engagement.
**Why it works:**
- Hybrid of consumption (boards followed) and contribution (pin saved).
- A user who saves a pin has signaled the product fits their use case.
**PM lesson:** for consumption + contribution products, the activation event often crosses both modes.
### LinkedIn: 5+ connections in the first week
LinkedIn's activation matches Facebook's pattern (network threshold within a window). 5 connections is enough to make the feed and "people you may know" useful.
**PM lesson:** repeating the network-threshold pattern. The exact count differs by product, but the shape is reliable.
### Figma: file created + 2+ collaborators invited
Figma's activation centers on collaborative file work. Single-user files are valid but not predictive of expansion to paid; multi-user files signal team adoption.
**PM lesson:** for B2B collaboration tools, activation is often the moment the file becomes a team artifact, not when it is created.
### Canva: design created + downloaded/shared
Canva activation is "user created a design and either downloaded it or shared it." The download/share signals real intent (not just exploration).
**PM lesson:** for creation tools, activation often pairs creation with output (download, share, export). Creation alone is too weak.
## Common patterns
| Pattern | Examples | When it applies |
|---|---|---|
| Network threshold | Twitter, Facebook, LinkedIn | Feed / social products |
| Volume threshold | Slack, Spotify | Engagement-driven products |
| Habit / frequency | Duolingo | Learning, habit-formation |
| Single-session "I get it" | Dropbox | Utility products with a clear "click" |
| Multi-object setup | HubSpot | B2B SaaS with structured data |
| Composite (X + Y) | Notion, Pinterest, Figma, Canva | Products with multiple value modes |
| Side-specific | Airbnb | Two-sided marketplaces |
## How to derive your activation event
1. **Pull two cohorts** from production data: users who retained at D30 (or your retention horizon) and users who churned.
2. **List candidate first-week behaviors** -- the 10-20 events most users do in week one.
3. **Compute the lift** -- for each event, what is the D30 retention rate of users who did it vs users who did not?
4. **Pick the highest-lift event** that is also reachable by most users.
5. **Tune the threshold** -- if the event is "messages sent", what count maximizes the discrimination?
6. **Validate** -- after picking the event, check that users who hit the threshold do retain at the predicted rate. If not, iterate.
## When the activation event needs to change
| Signal | Likely meaning | Action |
|---|---|---|
| Activation rate climbs but D30 retention does not | Activation event is no longer predictive | Re-derive against the latest cohort |
| Lots of users near but not at the threshold | The threshold is right at the inflection -- a small UX nudge has big payoff | A/B test threshold-adjacent flows |
| Activation rate is below 10% | Activation event is too hard | Re-evaluate; either lower the bar or improve the path |
| Activation rate is above 60% | Activation event is too easy | Re-evaluate; the event is probably not predictive enough |
## Anti-patterns
| Anti-pattern | Why it fails |
|---|---|
| "Completed onboarding" as activation | Onboarding completion is an output of UX, not a value signal |
| "Spent X minutes on the site" | Time is a noisy proxy; users tabbing away inflate it |
| "Created an account" | Signup is acquisition, not activation |
| "Clicked the upgrade CTA" | Intent signal, not value signal |
| "Used N features" | Feature usage is often product-led; users do not equate features with value |
| Activation event picked by intuition, not data | Will be wrong about half the time |
| Activation event copied from another product | Each product has its own value model |
## Reading list
- Sean Ellis + Morgan Brown, *Hacking Growth*, Chapters 4-5 (activation)
- Amplitude, *North Star Playbook* -- activation section
---
**Last Updated:** 2026-05-22
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |