20 posts a month is guessing. This is the engine.
JustVirals industrializes organic growth for app brands. We test formats at real volume, attribute every post to installs, signups and revenue, then extract the 3–5 formats that actually pay — ranked by the Rabk Rate Score.
- 600–1,500
- posts/month tested
- 30–50
- posts per format per week
- 3–5
- formats declared Viral DNA
- 70–90%
- output reallocated to winners
Why teams stall
Views are a diagnostic. Revenue is the verdict.
Most UGC programs fail on three things: not enough volume to be statistically real, no attribution from post to revenue event, and no rule for when to scale. The Engine fixes all three in the same loop.
Sample size
A format cannot be declared a winner below its minimum N — starting at 30–50 posts per format per week, stable across two windows.
Airtight attribution
Every asset carries format / angle / hook / CTA / creator tags plus deep links and UTMs, credited across D0, D1, D7 and D30.
Scaling discipline
Winners get 70–90% of output. Vanity formats — high views, low revenue — are flagged and cut on a stop-loss rule, not a feeling.
Signature metric
Rabk Rate Score — the credit score of virality that pays
One 300–850 number per format and per creator, rebuilt daily from your own attribution data. It decides what gets scaled, what gets tested, and what gets killed.
Format · Problem Receipt
812
Viral DNA Winner
Revenue power
40%revenue_per_post · revenue_per_1k_views · payback proxy
Conversion-chain efficiency
30%view→click · click→install · install→signup · signup→purchase
Volume & stability
20%N posts vs. minimum · week-over-week consistency
Quality & compliance
10%refund rate · brand safety · geo rules
The loop
Produce → measure → learn → refine → scale
Seven operational steps, run continuously. Nothing here is a one-off audit.
- 01
Diagnose
Tracking integrity, event schema, ICP → intent clusters.
- 02
Generate
10–20 candidate formats mapped to intent and objections.
- 03
Launch volume
600–1,500 posts/mo, every asset tagged at birth.
- 04
Measure
Daily ingest of performance + revenue events, D0/D7/D30.
- 05
Extract DNA
3–5 formats with proven revenue and stable chains.
- 06
Scale
70–90% output to winners, 10–30% to exploration.
- 07
Self-improve
Weekly retire/expand, monthly ICP + store refresh.
Dashboard
Rabk Rate leaderboards, not a wall of views
Rolling leaderboards by format, angle, hook, CTA and creator. Sorted by what moves money.
| Format | Rabk | Rev / post | Rev / 1k views | N (7d) | Verdict |
|---|---|---|---|---|---|
| Problem Receipt | 812 | $41.20 | $18.60 | 148 | Scale |
| Head-to-Head Teardown | 774 | $36.80 | $15.90 | 132 | Scale |
| Objection Autopsy | 731 | $31.40 | $14.10 | 96 | Scale |
| 60-Second Setup | 688 | $24.90 | $11.20 | 74 | Test |
| Contrarian Myth-Bust | 412 | $3.10 | $0.90 | 121 | Vanity |
Illustrative data. Your leaderboard is built from your own MMP + event schema.
Deliverables
Format cards a creator can shoot today
Each winner ships as a card: hook bank, timed script beats, visual structure, proof points, objection lines, CTA options, 8+ variants, failure modes and tracking tags.
Problem Receipt
Problem-aware · 'why is [task] so hard'
Open card →
Head-to-Head Teardown
Comparison · '[category] vs [competitor]'
Open card →
Objection Autopsy
Objection · 'is [product] safe / worth it / a scam'
Open card →
60-Second Setup
Tutorial · 'how to [outcome] fast'
Open card →
Contrarian Myth-Bust
Contrarian · 'is [common advice] wrong'
Open card →
Fail-safe · 30-Day Signal Plan
Not enough data yet? Then you get a plan, not a guess.
If volume or tracking can't support confidence, the Engine refuses to name winners and ships a 30-Day Signal Plan instead: posts per day required, 10–20 starter formats, the exact tracking spec, and a fixed decision date for extracting Viral DNA.