Method · published in full
How we review
A ranking is only worth something if you can see how it was built and disagree with it. Here is the whole model: four dimensions, their weightings, the tests behind each one, and the things we refuse to score.
- Dimensions scored4
- Heaviest weightingAccess, 35%
- Paid placementsNone
- Scores reviewedEvery 3 months
The starting position
Most exchange rankings are built for a user in the United States or the European Union. They weight things like instant bank transfers, domestic consumer protection and local currency pairs — all of which are either absent or irrelevant in Jamaica. A ranking built on those criteria will confidently recommend a platform that cannot take your money.
So we rebuilt the model around the Jamaican user's actual position: you hold a Visa or Mastercard from a Jamaican bank, your income is in Jamaican dollars or US dollars, no exchange lists a JMD pair, and no Jamaican authority regulates any platform you might use. Every weighting below follows from that.
The four dimensions and their weightings
| Dimension | Weight | What it measures |
|---|---|---|
| Access | 35% | Can a Jamaican resident actually onboard and fund? |
| True cost | 30% | All-in cost end to end, not the headline fee |
| Supervision | 25% | Named regulator, verifiable reference, reserve transparency |
| Exit | 10% | Getting value back into a Jamaican bank account |
Access — 35%
The heaviest weighting, because a platform that will not take your money scores zero regardless of how good its charts are. We test whether registration completes with a Jamaican address, whether identity verification accepts a Jamaican passport, driver's licence or national ID, whether cards issued by local banks are processed, and whether the platform publishes any country restriction that affects Jamaica.
We deliberately treat card acceptance as an access question rather than a cost question, because the difference between "expensive" and "impossible" is not a matter of degree. Where a platform's fiat permissions for Jamaican accounts are unclear — as they have been for at least one major venue — we mark it down and say so in the review rather than pretending to certainty.
True cost — 30%
We price a purchase end to end. That means the platform fee, the spread on the instrument used, and the foreign-exchange margin a Jamaican bank applies when converting JMD to USD — a cost that never appears on any exchange fee page and which is frequently larger than the fee itself.
We score cost at three sizes rather than one, because the ranking changes with amount. A wire fee that is trivial on US$5,000 is punitive on US$150, and a platform with the lowest published trading fee can be the most expensive outcome from Jamaica once you count how you have to reach it. Our fee page shows the worked examples.
Supervision — 25%
Not marketing language about bank-grade security. Three specific things: a named regulator, a licence or registration reference we can look up on that regulator's own register, and — where it exists — independently verified reserve reporting.
This weighting is higher for Jamaica than it would be for a market with domestic licensing, and that is deliberate. The Bank of Jamaica has stated that it does not regulate or supervise virtual currencies and has authorised no entity to operate a virtual currency platform. Until the Financial Services Commission's proposed Virtual Assets Service Provider regime commences, whatever supervision a platform has is the only supervision you get. A platform that cannot name its regulator has, in effect, failed the whole review.
Exit — 10%
The dimension people forget until they need it. Can you turn a balance back into money in a Jamaican bank account, how long does it take, what does it cost, and does the platform produce statements a Jamaican bank will accept when it asks where the funds came from?
It is weighted lowest not because it matters least but because the variation between reputable platforms is smaller. Where a platform has a history of withdrawal problems, that shows up as a penalty rather than a low sub-score. Our cashing-out guide explains why documentation is the part that matters most.
The tests we actually run
- Registration with a Jamaican address. Does the form accept it, and does the account open?
- Identity verification with a Jamaican document. How long, how many attempts, what is requested.
- A small real purchase. We record four numbers: the JMD debited, the USD credited, the asset quantity received, and the mid-market rate at that timestamp. That is the only honest way to measure all-in cost.
- A withdrawal of that same small amount. Back out to the account it came from. Every fake platform in existence fails this, and it is the single most informative test we run.
- A licence check. We look the reference up on the regulator's own register, reached by typing the regulator's address rather than following a link the platform supplied.
- A support question. One real question, asked through the platform's own channel, timed.
- Fee schedule reconciliation. Does the published schedule match what we were actually charged?
Where we cannot complete a test — because a platform will not onboard us, or because a route requires an amount we are not going to send — we say so in the review rather than inferring a result.
The test we would give every reader
Deposit a small amount and withdraw it straight back before you commit anything meaningful. It costs a few hundred Jamaican dollars in fees and it is the most informative thing you can do. Test number four above is the whole method in one action.
What we deliberately do not score
Number of tokens listed. Breadth of listings is a proxy for risk appetite, not quality. A platform with four thousand assets is not better than one with three hundred for someone buying Bitcoin monthly.
Promotional rates and sign-up bonuses. They expire, they are conditional, and they reward the platform with the biggest marketing budget rather than the best product.
Leverage and derivatives features. We do not treat access to 100x leverage as a positive for a retail Jamaican user, and in several reviews we treat prominent placement of it as a negative.
App store ratings. Gameable, and dominated by users in other countries with different access.
Staking yields. A yield is a return with a risk attached, and comparing yields without comparing the risks is how people end up lending assets to something that fails.
How commercial links are handled
Some links on this site are commercial partnerships and we may be paid when a reader opens an account through one. Three rules govern that.
The score is calculated before the commercial question is asked. A platform's sub-scores across the four dimensions are recorded from the test results; nothing about a partnership enters that calculation.
Every ranking carries its reasons. If you think our top-ranked platform is ranked too highly, the review states exactly which sub-scores drive it — published licences, local card acceptance, both an instant-buy route and a spot order book — so you can argue with the specifics rather than the conclusion.
Negatives are published for partners too. Our top-ranked review lists the drawbacks: no JMD trading pair, card convenience fees far above order-book pricing, and no physical presence anywhere in the Caribbean. A review that only praises is an advertisement.
How often scores change, and how to challenge one
We re-check fee schedules, country availability and licence references at least quarterly, and immediately when a reader tells us something has changed. Every table on the site carries the date it was last verified.
If you think a score is wrong, write to contact@jamaica-crypto.com and say which sub-score and why. We will re-run the relevant test. If you are right, we will change the score and note the change. If you work for an operator, that address is the right one and we read those emails — but we will not move a ranking without a reason we can publish.