The first processor with public freeze statistics and a guaranteed human path.
The score the risk system already keeps, on your dashboard every day — with the signals that move it and the thresholds it would have to cross.
Account risk score
review at 60 · payouts hold at 85 · you are at 18
recomputed daily · the same number the risk desk reads
your score · your signals · your headroom
Nothing happens and nobody is looking. The dashboard says so, in those words.
A named human reads the account. You see who, which signal tripped it, and your headroom back to normal.
Payouts pause and the notice below is what arrives — a stated reason, a release date, a reviewer.
No band is a secret. The thresholds sit in the published policy, and the score moves on your dashboard long before the band ever does.
published thresholds · no silent state
Every processor already computes this number. It just lives in a risk console the seller never sees. Putting it on the dashboard is the whole feature — a hold should never be the first time you hear about your own risk.
SIMULATED Simulated figure. Illustrative only — not a measurement.One account, one restriction, two ways of telling you. The difference is not tone — it is whether the notice contains facts.
no reason · no date · no name
Signals Pro crossed the chargeback threshold. Your other three products keep selling and keep paying out on schedule.
Held funds release on Sep 12, 2026 · 00:00 UTC
day 4 of 6 · the date was set the moment the hold started
appeal · case OD-4417
stated reason · fixed date · named reviewer
Nothing here is a new capability. The rule, the measurement, the reviewer and the release date all exist inside the risk system already. Publishing them is a product decision, not an engineering one.
SIMULATED Simulated figure. Illustrative only — not a measurement.Six numbers, published on the first of the month, whether or not they flatter us.
Freeze rate, 30 days
Two in a thousand. Active sellers restricted in a rolling month, posted on the 1st.
Hours to a stated reason
Before lunch. From the hold landing to a specific rule being named in writing.
Days to resolution
Down from 9.4. Suspension to a final decision, twelve months of month-end medians.
Appeals overturned
We were wrong 31% of the time. Published because a low number would mean the appeal is theatre.
Funds-hold days
The tail is the number that matters. How long money actually sits — not the policy maximum.
Human escalation
A person, not a position in a queue. First response on the escalation path, measured from the request.
12 months · median
| Month | swop_ (days) | Industry typical (days) |
|---|---|---|
| Oct | 9.4 | 12.8 |
| Nov | 8.1 | 13.1 |
| Dec | 7.6 | 12.4 |
| Jan | 6.2 | 12.9 |
| Feb | 5.9 | 13.4 |
| Mar | 5.1 | 12.6 |
| Apr | 4.4 | 12.2 |
| May | 4.1 | 12.8 |
| Jun | 3.6 | 13 |
| Jul | 3.2 | 12.5 |
| Aug | 2.9 | 12.9 |
| Sep | 2.6 | 12.7 |
simulated · illustrative — a live version would read from production
side by side
| Metric | swop_ on Whop rails | Industry typical | What it measures |
|---|---|---|---|
| Freeze rate | 2.1 per 1k sellers | 7–9 per 1k sellers | Active sellers restricted in a rolling 30 days. |
| Hours to a stated reason | 1.4 h median | 48–96 h | From the hold landing to a specific rule being named. |
| Days to resolution | 2.6 d median | 12.7 d | From suspension to a final, written decision. |
| Appeals overturned | 31% | not published | Suspensions reversed after review. Published either way. |
| Funds-hold days | 4 p50 · 11 p90 | 30 p50 · 90+ p90 | How long money actually sits, not the policy maximum. |
| Human-escalation SLA | 15 min median | no SLA | First response from a named person, not a queue position. |
simulated · the baseline column is an illustrative range, not a measurement of any named company
Three documents a seller can actually hold you to: the rules, the appeal, and the clock.
What triggers a review, in plain English. Four rules, four numbers, versioned in public. If a threshold moves, the changelog says so before it applies to you.
Every suspension comes with four things. A reason you can argue with, a person who signed it, a clock that runs whether or not anyone is looking, and one free escalation.
Holds end on a date, not a vibe. The release date is generated when the hold starts and shown in the first notice. Extending it takes a written reason on the same page.
Concept demo, stated plainly. swop_ is not a Whop product and this page is not a Whop page. Every threshold, median and percentage above was invented to draw the idea — a real version would read from production, and would be worth exactly as much as the day it first reported a bad month.