The same coin is simultaneously traded on different exchanges at different prices. This gap — the spread — is the source of profit for an arbitrageur: buy cheaper on one platform, sell more expensive on another. However, in practice, discrepancies rarely exceed fractions of a percent and exist for mere seconds. Manual trading loses here: by the time a trader spots the opportunity and places two orders, the window has already closed.
The Arbitron platform automates this process. The service monitors order books and funding rates on 20 exchanges in real time, independently placing both sides of the trade — the so-called "legs." Meanwhile, funds remain on the trader's own exchange accounts, eliminating risks associated with transferring assets to a third party.
Four factors that eat into profits
The raw spread shown by the scanner is merely the difference between two quotes. Four key factors "compress" it down to real profit: taker fees on both legs, order book depth for the required volume, slippage during execution, and the speed at which the spread itself decays. A full cycle of two-legged arbitrage involves four market orders, and the choice of exchange pairing is critically important, as platform fees vary dramatically.
Speed decides everything: 23 milliseconds to Binance
Arbitron is connected to every trading pair on 20 exchanges, amounting to about 10,000 order books simultaneously. The platform keeps the entire order book in memory, not just the top quote, allowing it to calculate the real execution price for volumes of $25, $100, $500, or $1000. Each account receives a dedicated AWS server with its own static IP, and the trader chooses the region themselves: Tokyo, Singapore, Frankfurt, or London.
Server location is not a whim but a necessity. A request to Binance from Tokyo takes about 23 ms versus 206 ms from Singapore. A delay of 200–500 ms from another continent means the order arrives at a book that has already completely changed. In my measurements, such a trade cost nearly 0.8% in slippage — several times more than the typical spread. Exchanges also limit the number of requests from a single IP, and when operating from a shared pool, all service clients consume the limit at once, which directly slows down execution.
Backtest instead of pretty numbers
The platform has two search tools. Signals show data in real time, while the scanner works fundamentally differently: the table displays not the momentary spread, but the result of a backtest with a 5-minute delay. The platform runs the last 8 hours of recorded quotes through the same two-threshold strategy that trading cards use. Each row is the outcome of a simulation where fees have already been deducted.
A key feature is accounting for the trader's personal fee schedule. The user specifies their VIP level and native token discount in the exchange key settings, and the breakeven threshold is calculated from that number. The backtest delay is also factored in: the trader selects an interval of 0.5, 1, or 2 seconds, and each signal is calculated at the worst price within that period. This filters out the main illusion of arbitrage scanners — attractive percentages that are impossible to reach.
Trading cards: automation with protection
Working in Arbitron revolves around trading cards. Each one links a pair of exchanges, a specific coin, and two spread thresholds: for opening and closing. The exit threshold is calculated together with the cost of reverse trades, which immediately factors in the price of reversing both positions. Without such an adjustment, the spread looks profitable on paper but in reality brings a loss.
The card works on the discrepancy in either direction: if the spread moves up, the platform opens in one configuration; if it moves down, in a mirrored one. Many bots are rigidly tied to a single scheme and miss half of the movements. Delisting protection is a separate point of pride for the platform: three layers of monitoring, including polling the official announcement APIs of Binance, Bybit, Bitget, and OKX, allow it to stop position accumulation and notify the trader 3–14 days before a contract is removed.
Keys, fees, and withdrawals
Arbitron operates on a non-custodial model. The API key has permission only for trading; withdrawal rights are not requested. Keys are stored using an "envelope" scheme with AWS KMS, each record is encrypted with the AES-256-GCM algorithm, and each user has their own encryption key — there is no single master password.
There are three pricing tiers: Scanner at $39 per month (data only), Trader at $99 per month (dedicated server, static IP, up to 10 strategies, 35% fee on profits), and Prime at $299 (readying for launch, 30% fee, up to 50 strategies). The fee is charged once a week and only on realized profit: losses are carried forward, and income that covers a previous deficit is not taxed. Cashback on trading fees applies across seven exchanges — from 30% to 60%.
My verdict: Arbitron is a professional tool that solves the main problem of arbitrage — speed and completeness of analysis. However, it is worth remembering that even the most sophisticated automation does not eliminate market risks, and one should start with careful configuration of thresholds and limits, not with chasing maximum profitability.