Method

How Opta works

We rank providers by what you actually receive, not what they advertise. Public data, timestamped, recomputed daily.

What goes into a ranking

The receive-amount that lands in your account is four numbers stacked:

Spot: The reference rate at the moment of quote.
Spread: The bid–ask gap on the executing surface.
Fees: Trading fee, withdrawal fee, network gas.
Time: How long the path takes end-to-end.

Each one is sourced and disclosed.

How we source rates

Every quote belongs to one of five classes, named by where the price comes from. Different surfaces price differently. We treat each as its own class so rankings stay honest.

CEX orderbook: What we see when we depth-walk a public orderbook.
Fiat corridor: Bank-to-bank rate cards from the corridor leaders.
DEX curve: On-chain quotes routed through 1inch and LI.FI.
P2P: Peer-to-peer offers, currently LocalCoinSwap only.
Benchmark: Reference rates, not executable. Wise sits here.

Key terms · · · · ·

DeeperRead the full architecture of each rail class

What we audit, what we don't

Every executable rail Opta tracks gets audited daily. We compare the rate the provider quoted against the rate users actually receive after spreads, fees, and slippage. The gap is variance, the single number that tells you how seriously to take an advertised price.

We deliberately exclude two things from the audit:

Wise: Wise is a benchmark, not a tracked execution. We use its rates to anchor "what a reasonable corridor rate looks like," but we don't audit Wise against itself.
P2P beyond LCS: LocalCoinSwap is the only P2P provider we audit today. Adding more P2P sources requires a separate trust signal we haven't built yet.

Variance is read in three bands:

≤ 0.5%tight execution (emerald)

0.5 – 1.5%expected slippage (amber)

> 1.5%flagged for review (red)

Every audit row is also marked ok / warn / exclude. See the per-corridor audit table on /accuracy.

LiveSee the live audit on /accuracy