← Console
selfActualAdmin
Confirmed data
Directional · best available signal
GA4 not yet instrumented · Substack + vault session logs as proxy
North Star 1 Confirmed
Do users care?
68% ↑ 11pp this month
of users who complete session 2 return within 7 days · based on vault session logs
North Star 2 Directional
Does conversation weight improve over time?
+1.8 avg weight gain · sessions 1→5
directional · weight not yet tracked per-session · proxy from session frequency + depth signals · instrument to confirm

User behavior distribution

How users actually engage · directional
Daily / compounding15 users 31%
Weekly / activating13 users 27%
1–2 sessions / exploring19 users 42%
Total users
47
↑ 8 this month
Sessions logged
312
↑ 24%
Avg session depth
7.3
→ stable
Substack → signup
2.6%
↓ 0.3pp

Session activity — 30 days

Vault session logs · confirmed

Population distribution

Ratio moving? That's the signal.
Behavior patterns

Users by engagement pattern

Session count · return rate · weight trajectory · behavior read
UserSessionsLast seen7-day returnWeight trendPattern
What separates them

Daily / compounding

What they have in common
  • → Completed profile setup in session 1–2
  • → Used a gear within first 3 sessions
  • → Had a "it knows me" moment early
  • → Return cadence: every 1–2 days

Weekly / activating

The activation gap
  • → Profile partially complete
  • → Haven't hit a compounding session yet
  • → Return cadence: 5–10 days
  • → Most likely to convert if re-engaged

1–2 sessions / exploring

The question
  • → Treating it like a smart chat interface
  • → Low or no profile setup
  • → No gear usage detected
  • → Need the "wait, it knows me" moment
Substack → selfActual funnel

The owned pipeline

Directional · Substack subscriber data + basic traffic proxy · GA4 will confirm
Substack subs
88
confirmed
Clicked to SA
~34
~39% · directional
Signed up
~12
~35% of clicks · dir
Session 2
~8
~67% of signups · dir
Compounding
~3
~38% of S2 · dir
The gap that matters: Substack → signup is working. Signup → compounding is the unlock. The "exploring" bucket is where Substack-sourced users are landing. Onboarding is the lever.
Post-level performance

Posts driving signups · Substack data

Confirmed open/restack data · signup attribution directional
Conversation weight
Per-session weight tracking
Weight isn't yet stored per session in the vault. When it is, this panel shows weight trajectory per user — the clearest signal that the known-operator machine is working.
Instrument to unlock

Depth proxy · directional

Session frequency as weight signal · higher cadence = higher investment
Compounding
8.8
Activating
5.4
Exploring
2.8
What we're watching for
Signal: working
Compounding users return within 48h without prompt. Session count accelerates in weeks 2–3.
Signal: watch
Activating users plateau around session 4–5. If they don't cross into compounding by session 8, they rarely do.
Signal: instrument
Per-session weight stored in vault → this whole pane becomes confirmed data. One schema change.