Metrics and KPIs
What we measure decides what we improve. This page is the public-facing hierarchy: the one north-star at the top, the supporting concepts below, the SLAs we commit to, and the reporting cadence we run. Specific internal targets and stage-by-stage planning numbers live in our internal docs.
The north star
Weekly real-viewer-earned-USD.
We measure the dollar-equivalent value of accruals credited each week to wallets that pass the "engaged real user" heuristic (see anti-fraud). One number, across all tiers, all streamers, all distributions.
Why this:
- Captures viewer value created. If this grows, viewers are getting paid for time spent.
- Filters out sybil — using only real-user accruals, not gross.
- USD-denominated, so price moves in BTV/streamer-tokens don't distort the trend.
- Weekly cadence is fast enough to react and slow enough to smooth noise.
Supporting growth metrics (concept)
| Metric | What it captures |
|---|---|
| Active streamers | Streamers running ≥ 1 distribution per week — the supply side of the loop |
| Active linked viewers | Wallets linked + staked + accruing in the last 30 days — the demand side |
| Cash-out rate | % of accrued value swapped to USDC/SOL within 30 days of claim — the swap-revenue lever |
| Paid conversion | % of active streamers who upgrade Free → Pro / Plus within 90 days — the SaaS revenue lever |
Specific healthy bands per growth stage are operational planning numbers (internal).
Supporting health metrics (public commitments)
These are the targets we hold ourselves to and publish in the quarterly transparency report:
- Real-user earning ratio —
real_user_earned / total_earned. Target ≥ 95%. Below 90% triggers a review. - Streamer 30-day retention — % of streamers who run a second distribution within 30 days of their first. We publish this number quarterly.
- Viewer 30-day retention — % of first-time claimers who claim again within 30 days.
- Identity-tier liquidity depth — median 2% market-impact depth across listed streamer-token pools. We publish a quarterly snapshot.
Operational SLAs
These are public commitments we underwrite:
| Metric | Commitment |
|---|---|
| Backend API p95 latency | < 250 ms |
| Backend availability | 99.9% (≤ 43 min downtime/month) |
| Time to first accrual after wallet link | < 1 accrual tick |
| Claim transaction success rate | ≥ 99% |
| Swap router quote latency | < 1.5 s (Jupiter aggregator dependent) |
| Bug bounty SLA — critical triage | < 4 h |
| Bug bounty SLA — critical fix | < 72 h |
| Customer support first-response | < 24 h (Free) / < 4 h (Pro) / < 1 h (Plus) |
Misses are disclosed in the next quarterly transparency report.
What we don't track as a north star
- Total BTV staked — vanity number that bounces with token price.
- Number of distributions — easy to game by spammy streamers.
- Twitch hours covered — proxy for nothing in particular.
- Total swap volume — useful for revenue forecasting but can grow from speculation that doesn't reflect platform health. We pair it with real-user earning ratio to filter.
Reporting cadence
| Cadence | What |
|---|---|
| Daily | Operational dashboard (latency, errors, RPC health, MongoDB health) |
| Weekly | North star + supporting growth metrics, leadership review |
| Monthly | Full revenue breakdown, paid conversion, streamer cohorts |
| Quarterly | Public transparency report — emission progress, fee revenue, sanctions screen counts (anonymized), bug bounty stats, retention publish, SLA-miss disclosure |
| Annually | Audit summary, full P&L, strategy refresh |
Quarterly transparency reports are published to this checkpoint site once production launches.