Here is a question most founders cannot answer precisely: if one company suspended your account tomorrow, what percentage of your revenue stops? For many small internet businesses the honest answer is "most of it," and the uncomfortable part is that this power is not hypothetical, it is written into the published policies you agreed to. This article is a worksheet: map your chokepoints, score your exposure, and buy it down deliberately. No scare stories required; the primary documents are enough.
What the policies actually say
We have read these documents for earlier articles; assembled, they describe one consistent structure across very different platforms:
- Your payment rail can pause payouts at any time. Shopify's account-holds documentation states every Shopify Payments account "is legally subject to the standard review process, which happens throughout the account's lifespan", reviews are not a signup event, they are permanent. Your store may keep selling while payouts stop.
- Reserves can be indefinite. Stripe's reserves FAQ commits to releasing reserve funds at period end "if they are not needed", and notes that "in some rare cases, a reserve may be required indefinitely."
- Card networks set hard behavioral thresholds. Per Stripe's monitoring-programs documentation, Visa's program flags accounts at a 0.5% dispute-and-fraud ratio and treats 1.5% as excessive; Mastercard's starts at 100 chargebacks a month at 1.5%. Outcomes range from mandated remediation to termination.
- The same shape governs reach, not just money. Google's sender guidelines attach hard numbers (a 0.30% spam-complaint ceiling) to whether your email arrives at all. Different asset, identical structure: published thresholds, unilateral enforcement, limited appeal.
Marketplace sellers will recognize the pattern, seller forums are full of suspension stories with slow, opaque appeals. We could not verify marketplace-specific policies or incidence numbers to our sourcing standard (the policy pages and forum threads are login-walled), so treat marketplace risk as the same structure by analogy, documented above for payments, email and search. The structural point stands on the primary documents alone: on every major platform, continued access is conditional, reviewable, and revocable, and the conditions are numeric.
The audit: thirty minutes, one table
List every chokepoint where a third party sits between you and either money or customers. For each, four columns:
| Column | Question | Example answer |
|---|---|---|
| Chokepoint | What can be switched off? | Stripe account; Gmail deliverability; marketplace listing; app-store presence; social account |
| Blast radius | % of revenue or lead flow that stops | "82% of revenue settles through this one processor" |
| Tripwires | Which published thresholds apply to you? | Dispute ratio vs the 0.5%/1.5% lines; spam rate vs 0.30% |
| Recovery path | If cut off today, what is the documented appeal, and what do you do in week one? | Usually: email support and wait, which is the finding |
Score it
Our rule of thumb, 0-10: start at the blast radius of your single largest chokepoint (82% → 8.2). Subtract one point for each genuine mitigation in place: a second processor warm, an exported customer contact list you own, an owned channel producing leads, cash to survive 60 frozen days. Above 6 after subtractions, platform risk, not competition, is plausibly the biggest risk on your books.
The mitigation ladder, cheapest first
- Own your data tonight (hours, free). Export customers, orders, and content on a schedule to storage you control. A suspension that takes your audience contact list with it is a different disaster from one that doesn't.
- Watch your own tripwires (hours, free). The thresholds are published; monitor your dispute ratio and spam rate like revenue metrics. Our articles on keeping the dispute ratio down and operating under processor risk are, in this frame, tripwire management.
- Keep a second rail warm (days, small cost). A backup processor with real volume, a second sales channel with real listings, warm, not theoretical, because approvals take weeks precisely when you don't have them.
- Build one owned channel (months, compounding). Email list, publication, community, the only distribution a platform cannot revoke, argued fully in Distribution Is the Moat.
- Diversify actual revenue (quarters). The end state: no single company's review queue controls a majority of your income. Slow, and the only complete fix.
Limitations
Policies quoted were opened July 23, 2026 and change; we found no verifiable statistics on suspension frequency, so nothing here estimates the probability of enforcement, only its documented mechanics and your exposure to it. The scoring rule is a heuristic for prioritization, not risk math. Platform dependence is also often the right trade early on, the platforms exist because they deliver customers; the argument is for measuring the trade, not refusing it.
The bottom line
You cannot negotiate the policies, but you can know your number. Run the table this week: biggest chokepoint, blast radius, tripwires, recovery path. Then climb the ladder one rung a month, data export, tripwire monitoring, warm second rail, owned channel, until no single review queue can switch off your livelihood.
Related: the email list is the only audience you own, the cheapest rung on the mitigation ladder.
Discussion
Sign in with Google or just a name. No email link, no password to remember.