Coinbase: crypto exchanges rating breakdown

Custody of funds: 8/10

Proof-of-reserves: does it exist, who attests it, does it include liabilities (a PoR without liabilities is close to meaningless), how often. Insurance fund: size, is the address public, and has it ever actually paid out. Hack history — and above all whether users were made whole. Licences by jurisdiction, and the enforcement record behind them. Every claim here is sourced to a primary document: a regulator’s register, a court filing, or the venue’s own published proof.

Scored 8 of 10 against a category median of 5, which places it 3rd of 28 among centralised exchanges on this criterion. At a 40% weight that is 1.20 points above the median contribution of the weighted total. The best score in the category is 9, the worst 1.

No proof of reserves of any kind — no Merkle tree, no published reserve addresses, no way for you to verify your balance is backed. What it offers instead is a different species of evidence: it is a Nasdaq-listed company whose accounts and internal controls are audited annually by a major firm and filed with the SEC. Stronger in some ways, weaker in one that matters — you are trusting an audited balance sheet rather than a proof you can check yourself. A commercial crime policy — and its own annual report warns that total customer assets are “substantially more than our corporate assets and available insurance”. The same filing keeps a risk factor stating that in a bankruptcy, custodially held crypto could be drawn into the estate and customers “treated as our general unsecured creditors”. Read that sentence twice; Coinbase wrote it, not us.

True cost of trading: 2/10

Measured by us, not read off a fee page. We walk the live order book and fill $1,000 / $10,000 / $100,000 of BTC/USDT, then add the venue’s published taker fee. That sum — fee + spread + realised slippage — is the effective cost. Separately: the markup on network withdrawal fees, and the hidden spread inside “convert” / “buy in one click” flows.

Scored 2 of 10 against a category median of 6, which places it 27th of 28 among centralised exchanges on this criterion. At a 16% weight that is 0.64 points below the median contribution of the weighted total. The best score in the category is 10, the worst 1.

Liquidity and execution: 5/10

Order-book depth within ±0.1% and ±0.5% of mid, measured from the live book. Uptime over 12 months from the status page. Behaviour under stress: were there outages on days the market moved more than 5%.

Scored 5 of 10 against a category median of 6, which places it 17th of 28 among centralised exchanges on this criterion. At a 11% weight that is 0.11 points below the median contribution of the weighted total. The best score in the category is 10, the worst 1.

Assets and networks: 6/10

Number of pairs, adjusted for quality: the share of listings delisted within a year. Which networks are actually supported for deposit and withdrawal — not merely “supports USDT”.

Scored 6 of 10 against a category median of 6.5, which places it 15th of 28 among centralised exchanges on this criterion. At a 9% weight that is 0.04 points below the median contribution of the weighted total. The best score in the category is 10, the worst 1.

A curated, conservative catalogue — fine for majors, limited beyond them

Fiat on/off ramp: 9/10

Which methods are actually available per country, their fees, and their limits per verification tier — read from the venue’s published terms and cited. Where a venue does not publish this, that opacity is itself the finding, and it scores accordingly.

Scored 9 of 10 against a category median of 5.5, which places it 2nd of 28 among centralised exchanges on this criterion. At a 11% weight that is 0.39 points above the median contribution of the weighted total. The best score in the category is 10, the worst 1.

Broad, well-documented rails across many countries, and US customer cash sits at insured banks. You pay for the convenience in the spread.

KYC and access: 7/10

Verification tiers and their limits, what is genuinely usable without verification, and the country deny-list — from the venue’s published policy, cited. A venue that will not tell you your withdrawal ceiling before you sign up is marked down for it.

Scored 7 of 10 against a category median of 6, which places it 8th of 28 among centralised exchanges on this criterion. At a 8% weight that is 0.08 points above the median contribution of the weighted total. The best score in the category is 9, the worst 1.

Mandatory verification with published levels; no anonymous tier is offered or implied. United States: All 50 states — nationwide since it launched in Hawaii in August 2024. European Union: MiCA CASP — CSSF (Luxembourg). Coinbase Luxembourg S.A., passported to all 27 member states.

Product, API and docs: 9/10

API documentation quality, whether rate limits are documented, whether a public status page with uptime history exists, and the API response latency we measured ourselves while pulling the order book.

Scored 9 of 10 against a category median of 5.5, which places it 1st of 28 among centralised exchanges on this criterion. At a 5% weight that is 0.17 points above the median contribution of the weighted total. The best score in the category is 9, the worst 4.

Excellent developer documentation with documented rate limits, and a public status page. The consumer product is the most approachable in crypto — which is exactly what the spread is buying you.