Our scoring model weights five measurable factors so rankings stay defensible and reproducible.
Fees & spreads (25%): live maker/taker schedules plus realised slippage on $1,000 orders.
Security record (25%): custody model, proof-of-reserves, past incidents and insurance.
Liquidity depth (20%): order-book depth within 1% of mid price on major pairs.
Fiat access (15%): supported currencies, deposit rails and withdrawal settlement times.
Support & UX (15%): ticket response times, KYC friction and app stability.
Scores are updated whenever a provider changes a fee or suffers a material incident. We publish weak points as openly as strengths.
How this page was verified
- Fees re-measured on live accounts, not copied from a marketing page.
- Rankings are never sold; placement does not move a score.
- Corrections are dated and logged in public.
Last verified: 30 Sep 2026
Frequently asked questions
Short, dated answers. Every figure links back to the review it came from.
Where does the fee data come from?
The published maker/taker schedule on the platform plus a $1,000 market order we place ourselves to measure realised spread. Where the two disagree we publish both.
How do you measure liquidity depth?
Order-book depth within 1% of mid price on BTC, ETH and one mid-cap pair, sampled three times a day over a week.
What disqualifies a platform?
Unresolved withdrawal blocks, undisclosed custody of client assets, or refusal to publish a fee schedule. A disqualified platform is removed from the ranking rather than scored low.
Compare platforms on numbers, not marketing
Reader-funded, weekly re-verified, no paid placements.
The rate changes before the headlines do
One email each Tuesday: the offers that changed, the fees we re-measured and the platforms we downgraded.
Support ChainSignal
Keep independent crypto research free. Every USDT gift helps.
USDT on TRC20 / BEP20 / ERC20 / Solana / Arbitrum / Polygon via our partner checkout.