Methodology
How we compare providers
Our comparison pages exist to help you make a decision, not to spin one. Here's exactly where every figure comes from, what we don't claim, and how to get an error fixed.
Where the numbers come from.
Public sources only
Competitor prices, request allowances, update frequencies and features are taken from each provider's public pricing pages and documentation — never from private testing of their paid accounts or hearsay.
Dated and re-verified
Every comparison page lists its sources with the date we retrieved them, and shows a last-verified date. We re-check pages on a regular cycle and whenever a provider announces plan changes.
Like-for-like plans
We compare comparable tiers: entry paid plan against entry paid plan, and each provider's fastest publicly available update frequency with its price. Where plans don't map cleanly, we say so on the page.
Our own figures, labelled
fastFOREX API's 21ms average response time is measured across our production API traffic. It is our own figure, not an independent head-to-head benchmark — see below for what's coming.
Latency benchmarking.
What we measure today, and the independent index we're building.
Today
Latency claims on this site are fastFOREX API production averages. We encourage you to run your own benchmark during the free trial — every migration guide includes a copy-paste script that measures both providers from your infrastructure, which is the only latency that matters for your app.
Planned: a public latency index
We're building a recurring benchmark: scheduled requests to each provider's comparable endpoint from multiple neutral cloud regions, reporting p50 and p95 latency over a rolling window, with raw data downloadable. When it launches, comparison pages will cite it directly.
Fairness rules
Same request pattern for every provider: identical currencies, no caching, no keep-alive advantage, free/entry tiers only, measured from regions where neither party hosts infrastructure.
Corrections
If any figure on a comparison page is out of date or wrong, contact us with a link to the source and we'll verify and correct it promptly — and note the correction date on the page.
Fair questions.
About comparing your own product to competitors.
Aren't these pages biased by definition?
Why should I trust your 21ms claim?
What happens when a competitor changes their pricing?
The best comparison is your own.
Get a free API key and benchmark us against your current provider this afternoon.