Copytrading is what happens when the question “who knows what they’re doing?” meets cheap APIs. The proposition is simple: pick a trader whose published track record you find plausible, allocate a portion of your account, and the platform mirrors that trader’s positions on yours. The professional gets a fee or a follower count; the follower gets exposure to a strategy they couldn’t execute themselves; the platform gets the rails. As a product category, it has reshaped retail investing more than any single technology since the move from phone trading to web brokerage in the late 1990s.
It is also, at the major platforms, structurally broken in a specific way the marketing copy does not address.
What the product is, fairly described
Copytrading — sometimes called social trading — automates the mirroring of trades. A retail user picks one or more “lead traders” from a public roster, sets a copy ratio (the proportion of the follower’s account that follows each signal), and the platform routes the lead’s positions to the follower’s account, scaled by the ratio. Performance statistics are published — usually win rate, drawdown, compound return, time-on-platform. The follower can pause, unfollow, or rebalance at any point.
The case for the model is genuine. It lowers the bar for accessing a professional’s judgement from “hire a portfolio manager and meet the minimums” to “click once and post the minimum lot.” It produces an unusually rich, public, real-time dataset of trader performance — much of it good, much of it not, all of it visible. And it pushes the follower toward thinking about exposure in terms of strategy properties rather than individual instrument tips, which is a structural improvement on the prior retail-FX norm.
The conflict of interest the marketing copy doesn’t mention
Read the corporate filings of any of the dominant retail copy-trading platforms and a single structural fact recurs: the platform, the broker, and the execution venue are the same entity. The follower clicks “copy.” The lead trader opens a position. The platform receives the follower’s replica order. The platform fills it — typically against its own book, with the spread set by its own desk. The platform also collects the swap, the inactivity fee where applicable, and the spread on the follower’s eventual exit. The platform is on every side of the trade except the follower’s.
That is the textbook definition of a structural conflict of interest. It is not unlawful — every retail FX/CFD broker operating B-book runs on it — and disclosures cover it in small print. But it sits awkwardly with the copy-trading marketing surface, which sells the follower on the idea that the platform’s incentive is aligned with theirs. The platform’s actual incentive is to grow the follower balance, the lead-trader count, and the daily volume — regardless of whether followers compound or churn.
The Naga / eToro / ZuluTrade pattern is the same pattern as the broker-as-counterparty problem that has dogged retail FX for two decades, with a social-feed overlay.
Why the structure persists
Two reasons. First, the unit economics of a B-book broker running its own copy-trading platform are extraordinarily favourable. The platform sees every follower’s flow, prices its own risk, and can hedge selectively. Spreads, swaps and inactivity fees compound at a rate the sell-the-broker-and-buy-the-platform alternative could not match.
Second, the regulatory perimeter is unhelpful at telling the two structures apart. A copy-trading platform that holds a MiFID licence is presumed reputable; one that doesn’t is presumed risky. But the licence covers the brokerage, not the platform’s economic relationship to the follower. The question that matters — whose incentives are aligned with whose — sits outside the licence test entirely.
The unbundled alternative
There is a structurally different way to build this. Separate the introducer (the platform that selects, ranks and recommends lead traders) from the operator (the asset manager who actually executes the mandate). The follower contracts with the asset manager, which holds its own MiFID authorisation and executes through unaffiliated venues. The introducer is compensated for matching followers to managers whose track records hold up; the asset manager is compensated for compounding the mandate. No party with discretion over the follower’s trades is also the counterparty to those trades.
The 2017 CJEU ruling in Khorassani v Pflanz(Case C-678/15) is what makes this model practical: it places the brokerage of asset-management mandates outside MiFID’s authorisation perimeter, allowing the introducer to operate without a costly investment-firm licence while the manager retains the licence the manager needs. We’ve written about the ruling in detail here.
What followers should ask before signing up
Three questions, in order:
- Where is the spread?If the platform sets the spread on the follower’s trade, the platform has a per-trade incentive that runs against the follower’s compounding interest. That isn’t fatal, but it should be priced into the follower’s expected return.
- Who is the counterparty?If the platform is on the other side of the follower’s trade, every loss-on-trade is a profit to the platform. Read the execution disclosure.
- What does the platform earn when the follower loses? The honest answer at a B-book broker is: some of it. The honest answer at an unbundled introducer- plus-manager structure is: nothing. The difference is the alignment.
A note on the lead-trader side
None of the above is an attack on lead traders themselves. The professionals publishing track records on copy-trading platforms are, by and large, doing legitimate work. The problem is not the lead trader; it is the platform architecture that sits between the lead trader and the follower. Move the lead trader to an asset-manager mandate structure and the architecture problem goes away, but the professional’s skill does not.