Activation Funnel Diagnostic
by @quochungto
Use this skill to diagnose where in an activation funnel users drop off and decide between removing friction or adding 'positive friction' (guided steps) to...
Example 1: SaaS Tool with Empty-State Problem
Situation: A B2B analytics tool has 1,200 users sign up per month. Only 180 (15%) reach the aha moment (generating a first report). The team has funnel metrics but no survey data.
Process summary: 1. Aha moment confirmed: "user generates and views first analytics report" 2. Funnel metrics read: account creation (78%), workspace setup (61%), first data connection (42%), first report generated (15%) 3. Activation flow read: "first data connection" requires users to input API credentials or upload a CSV β no sample data available 4. Channel-segmented table built: paid search channel drops from 61% to 28% at "first data connection"; organic drops to 48% 5. Highest drop-off: "first data connection" β 490 users lost, 42% completion rate; paid search 2.2Γ worse than organic 6. Structural inference (no survey data): users are asked to provide credentials before seeing any product output; the product looks empty until connected; users arriving from paid ads may have lower intent than organic 7. Friction decision: add positive friction β the product requires users to set up before experiencing value; offer a sandbox dataset so users can generate a sample report before connecting real data 8. Experiments ranked: (1) add sample dataset for demo report β low effort, fast signal; (2) add progress bar showing "one step away from your first report" β low effort; (3) simplify API credential input form β medium effort; (4) add short video showing a completed report at the connection step β medium effort
Output:
activation-funnel-diagnosis.md: confirms empty-state as root cause, paid channel mismatch, positive-friction recommendationactivation-experiment-candidates.md: 4 experiments ranked by effortExample 2: Consumer App with Mid-Funnel Drop
Situation: A recipe and grocery app has 8,000 weekly installs. Funnel: app open (100%), browse items (72%), add to cart (48%), enter payment info (31%), first purchase (19%). Team has exit survey responses from users who reached the cart but did not purchase.
Process summary: 1. Aha moment confirmed: "user receives first grocery order as expected" 2. Funnel metrics read: steepest absolute drop is "add to cart β payment info" β 17% of all installs lost (1,360 users/week) 3. Activation flow read: payment info step requires new credit card entry and delivery address; no saved defaults; no indication of delivery fee until checkout summary 4. Channel-segmented table built: referral channel activates at 38% vs. paid social at 11% β large gap at payment step 5. Survey data analysis: top cluster (41% of responses) β users did not know whether delivery was free; second cluster (28%) β users forgot their first-order discount code 6. Friction decision: remove friction β users want the product (browse and cart rates are solid); specific information gaps are causing abandonment, not lack of understanding 7. Experiments ranked: (1) display delivery fee and first-order discount code automatically on cart page β low effort, addresses top two survey clusters; (2) simplify payment form with single sign-on (Google Pay/Apple Pay) β medium effort; (3) add delivery fee estimate earlier (browse screen) β low effort
Output:
activation-funnel-diagnosis.md: payment-step friction identified, two specific causes from survey data, remove-friction recommendationactivation-experiment-candidates.md: 3 experiments ranked, first two directly address surveyed reasons for abandonmentclawhub install bookforge-activation-funnel-diagnostic