Growth · this week vs last week
is the platform growing?
New users · last 30 days
creators = new broadcaster accounts · viewers = first web action (RSVP, vote or reaction)
New creators / day
New viewers / day
New-creator funnel · accounts created in the last 90 days
how far new creators get · where they stall · what broke on the way
Creators reaching each stage
tracked accounts only
Stages
stuck = stopped before this stage and quiet 3+ days · median = time from the previous stage
| Stage | Reached | % | Stuck here | Median |
|---|
Stuck creators
most recent stalls first · ⚠ = something broke for them (tried and failed, not just never tried)
| Creator | Stuck at | Last step | Quiet for | What broke |
|---|
What broke
failures recorded against new creators, by the stage they happened in
| Failure | Stage | Creators | Times | Last |
|---|
Guided tour
of the creators each step was shown to, how many got past it · dismissed = pressed “Got it” (asked, and said not now — they may still do it later) · ⚠ marks the step most creators saw and stopped at
| Step | Shown | Got past it | % | Stalled | Dismissed |
|---|
Recent activity
the latest across the platform
🗓 Recent events
newest events created
👤 New users
newest signups
💜 Recent bits
each tip · who · which event
🗳 Recent votes
each vote · who · which event
Explore · last 14 days
what Explore showed · did anyone open the event, then go, vote or react within a day?
Impressions / day
cards shown on Explore
Share-page visits by source
unique visitors per event, by where they came from
Rails
impression → clicked (opened the share page) → converted (going / voted / reacted) → attended (said going, event ran) · 24h attribution
| Rail | Impressions | Clicked | CTR | Converted | CVR | Attended |
|---|
Promotions
running promotions · and the promotion-credit ledger behind them
Running now
placements = the served counter (slots filled, not views) · logged = promoted impressions in the Explore log
| Event | Creator | Started | Ends | Placements | Logged | Budget |
|---|
Recent credit movements
every grant and spend · newest first
| When | Creator | Change | Why | Event |
|---|
Points · earned currency · 250 points buy one promotion credit
earning, spending, and challenges
Earned vs spent / day
last 30 days
Where points come from
all-time points earned, by action
Top balances
| Holder | Balance |
|---|
Beta list
who left an email · and who agreed to be mailed
Reach
how many people actually turned up — and did reminders help?
Creators
are people scheduling, going live, and finishing?
Create → Live → Complete
funnel across all events
Scheduling lead time
how far ahead events are scheduled
Top creators
🌟 thriving (real engagement) · ✅ healthy · 👻 hollow (empty events) · ⚠ struggling · 🌱 new
| Creator | Created | Went live | Missed | Go-live rate | Bits earned |
|---|
Revenue · broadcasters raise bits → OneStarLive earns 20% · 1 bit ≈ $0.01
Your cut over time
dev-cut USD per day · last 30 days
Top earning streamers
who EARNS the most (real streamer name)
| Streamer | Gross bits | Total $ | Your cut |
|---|
Goal types
what kind of events people run
Community
celebrate your top viewers
Top spenders
most bits
| Viewer | Bits |
|---|
Top reactors
most reactions
| Viewer | Reactions |
|---|
Top voters
most votes
| Viewer | Votes |
|---|
Top attendance
shows up most
| Viewer | Attended |
|---|
Extension health
who has the Twitch extension active · and are they using it?
Activity tier
hot / warm / cold / dormant
Active vs idle
ext-active creators: made an event in 30d?
Live performance · real-time poll-endpoint health (resets on restart)
viewer poll pressure, latency, and whether the scale cache is working — refreshes with the dashboard
Logs & errors
broadcaster-facing failures & automation delivery · real bugs surfaced, expected conditions filtered
Broadcaster action failures · last 14 days
Create Clip / Marker / Set Stream Info / lifecycle — grouped by what failed & why
| What failed | Why (reason) | Count | Severity | Most recent · event |
|---|
Twitch EventSub health
Dead subscriptions deliver nothing and never recover on their own — the subscribe path now self-heals them, this is the proof
| Type | Enabled | Pending | Dead |
|---|
| Broken id | Type | Status | Count |
|---|
automation delivery failures by lifecycle stage · red herrings filtered
Automation failures by stage & error
real bugs vs. expected/benign
| Stage | Error | Count | Severity | Meaning / likely cause |
|---|
Trends over time
daily history · last 30 days
New events / day
Events by status
current lifecycle state
Automation delivery
sent · failed · skipped by destination