PLG Motion Designer

A product-led motion lives or dies on one number most teams never define: the percentage of signups who reach value before they churn in silence. You design the path from signup → aha moment → activated and instrument the gates between each step, so the product sells itself instead of waiting on a human. The most common failure is a vanity definition of activation - "created an account" - that correlates with nothing. This skill fixes the definition first, then builds the motion around it.

When to use this skill

Reach for it once positioning and the offer are set ([[positioning-statement]], [[messaging-hierarchy]]) and the self-serve product can be signed up for without a sales call. It owns the in-product journey; the acquisition page is [[landing-page-copy]], the paywall economics are [[saas-pricing]] / [[pricing-strategy]], and the broad launch plan is [[go-to-market-planner]] and [[launch-plan-sequencer]]. Run it before [[launch-day-runbook]] so the motion is live and measured when traffic hits.

Step 1 - Define the aha moment off behavior, not opinion

The aha moment is the first in-product action after which a user's odds of retaining jump sharply. Find it, do not guess it:

1. List the 8-12 candidate first-week actions a user can take.
2. For each, compare week-4 retention of users who did it vs. didn't. The aha action is the one with the widest retention gap that a new user can plausibly reach in the first session.
3. Pin it to a number, not a vibe: not "sends a message" but "sends a message to a second person." Slack's was ~2,000 messages sent by a team; the shape (frequency x breadth x time) matters more than the figure.

A crisp aha definition reads: "<count> of <core action> across <breadth> within <time window>." If you cannot fill that template, you do not yet know your aha moment - and every downstream nudge is guesswork.

Step 2 - Reverse-engineer activation milestones to it

Activation is reaching the aha moment. Work backward and name the 3-5 ordered milestones a user must clear to get there. Each milestone is a discrete, logged event - the rungs of the ladder:

  M0  Signup            account created
  M1  Setup             the one config without which value is impossible
  M2  First action      core action done once (the "empty state" defeated)
  M3  Aha moment        the Step-1 definition is met  ← ACTIVATED
  M4  Habit / PQL       repeated value → ready to convert (hand to saas-pricing)

Rules that keep the ladder honest:

• Strip every step that is not on the critical path to value. Each extra step leaks users.
• Order them; a milestone a user reaches out of order is mislabeled.
• M2 is the empty-state killer - the single hardest drop, where a blank product meets a new user. Design it deliberately (templates, sample data, a guided first action).

Step 3 - Gate each step with a metric

Every transition between milestones gets two numbers, and you do not move on until they are instrumented:

• Step conversion rate - % who advance Mₙ → Mₙ₊₁. This localizes the leak.
• Time-to-value (TTV) - median time from signup to M3 (aha). The headline speed metric; long TTV silently kills self-serve.
• Activation rate - % of signups who reach M3 within the TTV window. The one number the whole motion optimizes.

The biggest single drop in step conversion is your bottleneck - fix it before touching anything else. Optimizing a 90% step while a 30% step bleeds upstream is the classic waste.

Step 4 - Place in-product nudges only on the gaps

Nudges are interventions to lift a specific step's conversion - never decoration. For each leaking step, pick the lightest nudge that moves it:

Discipline: one nudge per leak, instrument its effect, keep it only if step conversion rises. Over-nudging (badges, popups, tours everywhere) trains users to dismiss everything and buries the one that matters.

Runnable artifact - activation funnel diagnostic

Drop your per-milestone counts in; it returns step conversion, where the funnel leaks worst, TTV, and the overall activation rate. Self-contained Python.

# activation_funnel.py  -  python3 activation_funnel.py
from statistics import median

# Ordered milestones: signup is M0, aha is the activation gate.
MILESTONES = ["M0 signup", "M1 setup", "M2 first action", "M3 aha (ACTIVATED)"]
COUNTS     = [10000,        6200,        3900,             2600]   # users reaching each
AHA_INDEX  = 3                                                     # index of the aha milestone
TTV_HOURS  = [0.5, 6.0, 22.0, 34.0, 48.0, 70.0]                   # signup→aha per activated user

def report():
    print(f"{'transition':<34}{'reached':>9}{'step conv':>11}")
    worst = ("", 1.0)
    for i in range(1, len(MILESTONES)):
        conv = COUNTS[i] / COUNTS[i - 1]
        arrow = f"{MILESTONES[i-1].split()[0]} -> {MILESTONES[i].split()[0]}"
        print(f"{arrow:<34}{COUNTS[i]:>9}{conv:>10.0%}")
        if conv < worst[1]:
            worst = (arrow, conv)
    activation = COUNTS[AHA_INDEX] / COUNTS[0]
    print("-" * 54)
    print(f"activation rate (signup -> aha){'':<3}{activation:>20.1%}")
    print(f"median time-to-value (hours){'':<6}{median(TTV_HOURS):>20.1f}")
    print(f"biggest leak{'':<22}{worst[0]:>16}  ({worst[1]:.0%})")

report()

Worked output for the sample numbers above:

transition                          reached  step conv
M0 -> M1                               6200        62%
M1 -> M2                               3900        63%
M2 -> M3                               2600        67%
------------------------------------------------------
activation rate (signup -> aha)                  26.0%
median time-to-value (hours)                      28.0
biggest leak                          M0 -> M1  (62%)

Read: 26% activation with the worst leak at the very first setup step (M0→M1) and a 28-hour TTV. The move is a Step-4 nudge on setup (inline checklist) and a hard look at why first value takes a day - not a new feature, not more signups.

Fill-in template - the activation spec

Aha moment:    <count> of <core action> across <breadth> within <time window>
Activation =   reaching the aha moment within <TTV window>

Milestones (ordered, each a logged event):
  M0 Signup        → event: ______
  M1 Setup         → event: ______   gate: step conv ____%  nudge: ______
  M2 First action  → event: ______   gate: step conv ____%  nudge: ______
  M3 Aha/ACTIVATED → event: ______   gate: step conv ____%
  M4 Habit / PQL   → event: ______   (hand to saas-pricing for conversion)

North-star gate:   activation rate = ____%   (target: ____%)
Speed gate:        median TTV = ____         (target: ____)
Current bottleneck: step ______ at ____%  →  intervention: ______

Quality bar

• The aha moment fits the <count> of <action> across <breadth> within <window> template and is backed by a retention gap, not an opinion.
• Every milestone is a discrete logged event, ordered, on the critical path to value - no decorative steps.
• Each transition has a step-conversion number and the whole motion has an activation rate and a median TTV. If a step is uninstrumented, the motion is not done.
• Exactly one nudge sits on each leaking step, and each nudge has a measured before/after on that step's conversion.
• The single biggest leak is named, and the recommended action targets it - not a healthier step.
• M4/PQL hands cleanly to [[saas-pricing]] / [[pricing-strategy]] for conversion; this skill stops at activation.

Do NOT

• Do NOT define activation as "signed up" or "completed onboarding tour." Those correlate with nothing; activation is reaching value.
• Do NOT optimize a healthy step while an upstream step bleeds. Fix the biggest leak first.
• Do NOT carpet the product in tours, badges, and popups. One nudge per leak, kept only if it moves the number.
• Do NOT use this for sales-led or outbound launch sequencing - that is [[launch-plan-sequencer]] and [[launch-day-runbook]].
• Do NOT use this to arm a human sales team with decks, one-pagers, or objection handling - a PLG motion replaces the rep; for the sales-led path use [[sales-enablement-kit]] instead.
• Do NOT design the signup/acquisition page or the paywall here - that is [[landing-page-copy]] and [[saas-pricing]] / [[pricing-strategy]].
• Do NOT invent the aha-moment number to look good in a deck. An unbacked activation target sends the whole motion chasing the wrong behavior.
• Do NOT confuse a long TTV with a feature gap; it is usually a leaking step or a missing nudge, found in Step 3.

Deliverable

A one-page activation spec: the aha-moment definition with its retention backing, the ordered milestone ladder, the step-conversion / TTV / activation-rate gates, one nudge per leaking step with its measured lift, the named current bottleneck and the intervention for it, and a clean handoff at M4/PQL to [[saas-pricing]] for conversion. Fits inside the GTM launch built by [[go-to-market-planner]] and [[launch-plan-sequencer]].