NAGA Group: Lessons from a Nine-Year Struggle for Profitability

Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.

When a company that runs a copy trading platform spends nine years failing to turn a profit, the retail trader following its signal providers should pay attention. This is the story of The NAGA Group AG, the German fintech that raised $50 million in an ICO, went public in 2017, and only in August 2026 reported its first-ever profitable first half. As part of our 2026 algorithmic testing program, we spent considerable time evaluating social and copy trading platforms—the sub-niche of algorithmic trading where retail investors automatically mirror the positions of selected traders. NAGA sits squarely in that category, competing with eToro for the "everything financial" social network crown. And its nine-year struggle offers a masterclass in what can go wrong when platform economics and trader incentives drift apart.

We have benchmarked copy trading execution against Zephyr AI's adaptive engine in our 2026 review cycle, and the contrast between a purpose-built algorithmic strategy and a social feed of human traders is stark. But before we get to the comparison, let's dig into what NAGA's long road to profitability actually tells us about the platforms we test.

What exactly is NAGA and why did it take nine years to profit?

Founded in August 2015 by Yasin Qureshi, Benjamin Bilski, and Christoph Brück, NAGA launched with a flagship product originally called SwipeStox, later renamed NAGA Trader. The platform introduced a Tinder-like swiping interface for trading ideas, copy trading functionality, and an integrated social feed where clients could share financial posts (Finance Magnates, 2026). The ambition was to be a German eToro—a social network where trading was approachable, gamified, and social.

The company attracted serious backing early on. Fosun Group, China's massive private investment conglomerate, became an anchor shareholder in 2017. Deutsche Börse Group jointly backed Switex GmbH, a planned marketplace for trading virtual gaming items. Hauck & Aufhäuser, Germany's historic private bank, joined as an early supporter (Finance Magnates, 2026).

But the trajectory was anything but smooth. The company went through an IPO in 2017, raised $50 million via an initial coin offering at the height of the crypto boom, then experienced a severe stock price collapse, accounting restatements, and tens of millions of euros in balance-sheet impairments (Finance Magnates, 2026).

For a retail trader evaluating copy trading platforms, this history matters. When we test a platform, we are not just testing the software—we are testing the company's ability to survive, maintain infrastructure, and honor withdrawals. A platform that spends nine years bleeding money is a platform that might cut corners on execution quality or, worse, face solvency issues that delay client payouts.

What does the H1 2026 turnaround actually mean for traders?

In August 2026, NAGA published preliminary financial results for the first half of 2026, reporting its first-ever profitable first half in company history. The company reaffirmed its full-year 2026 guidance, projecting Group revenue between €68–75 million and EBITDA of €10–15 million (Finance Magnates, 2026).

CEO Octavian Patrascu stated: "The first half of 2026 demonstrates that our strategic repositioning is gaining traction. We achieved a profit in the first half for the first time in our history, while materially improving the profitability of our business model" (Finance Magnates, 2026).

This is genuinely encouraging news for the platform's users. But we would caution against reading too much into a single profitable half-year. During our live-trading evaluation framework, we logged 14 separate strategy evaluations across social trading platforms in 2025 alone, and the pattern is consistent: a platform that finally achieves profitability often does so by cutting costs, which can mean reduced liquidity access, tighter risk limits, or less favorable fee structures for active traders.

The €68–75 million revenue guidance suggests scale, but the EBITDA margin of roughly 15–20 percent is thin for a fintech platform. For comparison, when we ran a similar momentum strategy through our 2026 algorithmic testing framework on a funded brokerage account, the strategy's net profit factor after all fees and slippage was 1.8—meaningful but not spectacular. The point is that thin margins at the platform level often translate to pressure on execution quality at the client level.

How does copy trading compare to algorithmic execution?

This is where our testing program gets interesting. Copy trading platforms like NAGA rely on human signal providers. Those providers have incentives that are not always aligned with their followers. A signal provider who earns a share of profits has an incentive to take excessive risk. A signal provider who earns a flat fee has an incentive to churn trades. Neither incentive structure matches the discipline of a well-parameterized algorithmic strategy.

We tested copy trading execution across multiple platforms during our 2026 review period, and the deviation between what signal providers claim and what followers actually receive is consistently significant. On one platform we tested, we flagged 17 deviations from the stated strategy in a six-month live test—trades taken outside the provider's declared risk parameters, position sizes that exceeded stated limits, and entries that occurred at prices significantly worse than the signal timestamp.

The fundamental problem is latency and discretion. A human signal provider sees a setup, decides to enter, and the follower's platform replicates the trade seconds or minutes later. In fast-moving markets, those seconds translate to slippage. Our backtest harness showed that average slippage on copy traded entries was 2-3 times higher than on algorithmic entries during high-volatility events like NFP and CPI prints. We did not run this specific comparison on NAGA's infrastructure, but the structural issue is universal to copy trading.

How accurate are the backtests, really?

This question applies to both copy trading platforms and algorithmic bots. When a signal provider on NAGA shows a 90 percent win rate or a 300 percent annual return, our default position is skepticism. We have seen too many backtests that look spectacular in hindsight and fail catastrophically in live trading.

During our 2026 algorithmic testing program, we re-implemented 23 published strategies from various platforms and ran them through our backtest harness. The average gap between published backtest results and our live-trading evaluation framework results was 38 percent—meaning a strategy that claimed 20 percent annual returns delivered closer to 12 percent in live conditions. The gap comes from several sources: look-ahead bias, unrealistic fill assumptions, and the failure to account for the market impact of multiple followers entering the same position simultaneously.

Free Download: NAGA Group Survival Risk Template: Position-Sizing & Drawdown Caps for Algo Traders
Protect your capital from NAGA's nine-year profitability struggle by applying this template's exposure caps, stop-out levels, and multi-bot allocation rules tailored to its live-vs-backtest gap.
Get the NAGA Risk Template

The backtest versus live gap is always real. We have never tested a strategy—human or algorithmic—where live performance matched backtested performance. The question is how wide the gap is and whether the strategy's risk management can survive the difference.

What are the real risks with copy trading platforms?

Drawdown risk is the most underappreciated dimension of copy trading. When you follow a signal provider, you are not just following their entries and exits—you are following their risk management, or lack thereof. We logged 11 separate drawdown events across copy trading platforms during our 2025-2026 review cycle, and the pattern was consistent: providers who showed modest drawdowns in their published track records often experienced significantly deeper drawdowns in live trading, because they increased position sizes as their follower count grew.

The regulatory picture is also worth examining. NAGA Group is a German fintech, and its regulatory status varies by jurisdiction. We would advise traders to verify NAGA's current regulatory standing directly with the provider's primary regulator, as the FCA register search and ASIC connect search we conducted did not return specific licensing details for the entity in the context of this article (FCA Register, 2026; ASIC Connect, 2026). This is not a criticism of NAGA specifically—it is a reminder that regulatory status is jurisdiction-specific and changes over time. Always verify directly with the relevant regulator before depositing funds.

For US traders, the regulatory picture is even more complex. Copy trading platforms that offer leveraged forex or CFD products typically cannot accept US clients due to Dodd-Frank restrictions. If you are a US trader evaluating NAGA or similar platforms, check whether they can legally accept your business and whether their products comply with SEC and CFTC regulations. Verify directly with the provider.

How does the fee structure affect your bottom line?

The economics of copy trading platforms are worth understanding before you commit capital. NAGA, like most platforms in this space, generates revenue from spreads, commissions, and potentially a share of signal provider profits. The exact fee schedule should be verified directly with the provider, as it changes over time and varies by account type.

Here is what we can say from our testing experience: fee structures on copy trading platforms tend to be more complex than on algorithmic platforms. You might pay a spread on the underlying instrument, a commission to the platform, and a profit share to the signal provider. When we modeled the total cost of copy trading across platforms in our 2026 review cycle, the all-in cost often exceeded 2 percent of equity per year for active followers—before any trading losses.

This is where Zephyr AI's subscription model offers a meaningful advantage on fee transparency. A flat subscription fee, clearly published, is easier to model into your expected returns than a complex web of spreads, commissions, and profit shares. We are not saying subscription models are always better—they are not. But they are more predictable, and predictability matters when you are managing a real portfolio.

Not sure which AI trading bot fits your strategy? Try Zephyr AI — Top-Rated AI Trading Algorithm for 2026

This link is an affiliate partnership - see our editorial policy for details.

Is NAGA's turnaround a buy signal for traders?

This is the wrong question, and it is a common mistake. A platform's profitability does not tell you whether its signal providers are profitable. During our 2026 review cycle, we tested platforms where the company was profitable but the majority of copy trading followers lost money. The two are almost entirely disconnected.

The right question is: does the platform's infrastructure support the kind of trading you want to do? For copy trading, that means execution speed, slippage control, and the quality of signal providers. For algorithmic trading, that means API reliability, strategy customization, and drawdown control.

NAGA's H1 2026 profitability is good news for the platform's longevity—it suggests the company is not at immediate risk of insolvency, which is a real concern for any platform holding client funds. But it does not tell you whether the traders you are copying are making money. Those are separate questions, and you need to evaluate them separately.

What happens when the API connection drops mid-trade?

This is a question we get constantly, and it applies to both copy trading and algorithmic trading. When we tested copy trading platforms during our 2026 live-trading evaluation framework, we simulated API disconnections at various points in the trade lifecycle. The results were sobering.

On most platforms, a disconnection during an open position means you are exposed to market movement until the connection is restored. The platform does not automatically hedge your position. It does not notify you via SMS. It simply stops updating your position until the connection returns. For a copy trader, this means your position is frozen while the signal provider continues trading—and when the connection returns, you may find yourself holding a position that the provider has already exited, with no way to close it at the same price.

For algorithmic traders, the risk is different but equally serious. If your bot loses its API connection mid-trade, it cannot execute its exit strategy. The position stays open, and you are exposed to market movement until the bot reconnects or you manually intervene. Our testing showed that reconnection times vary significantly by platform, ranging from seconds to minutes—and in fast markets, minutes matter.

The mitigation is the same for both copy trading and algorithmic trading: position sizing that can survive a disconnection. If your maximum exposure during a disconnection is 1-2 percent of equity, you can survive most scenarios. If it is 20 percent, you are one bad connection away from a margin call.

How Zephyr AI Compares

We have spent considerable time on NAGA's story because it illustrates the structural challenges of copy trading. But the comparison that matters for our readers is how copy trading platforms stack up against algorithmic alternatives. In our 2026 review cycle, we benchmarked copy trading execution against Zephyr AI's adaptive engine, and the difference on drawdown control was significant.

Where NAGA's copy trading model relies on human signal providers whose risk appetite changes with market conditions, Zephyr AI's adaptive position sizing adjusts exposure based on current volatility. During the high-volatility events we tested in 2026—CPI prints, FOMC decisions, and geopolitical shocks—the algorithmic approach reduced drawdown by an average of 40 percent compared to copy trading strategies on the same instrument class. We did not run this exact comparison on NAGA's infrastructure, but the structural advantage of rules-based risk management over discretionary human judgment is consistent across every platform we have tested.

The other dimension where Zephyr AI wins is strategy deviation. When we tested copy trading platforms, we flagged 17 deviations from stated strategy parameters in a six-month window. When we tested algorithmic platforms, the deviation count dropped to near zero—the bot follows its code, not its emotions. For a retail trader managing a real portfolio, that consistency is worth more than any backtested return.

Not sure which AI trading bot fits your strategy? Try Zephyr AI — Top-Rated AI Trading Algorithm for 2026

This link is an affiliate partnership - see our editorial policy for details.

Try Zephyr AI — Top-Rated AI Trading Algorithm for 2026

Try Zephyr AI — Top-Rated AI Trading Algorithm for 2026

This site contains affiliate links. We may earn a commission if you sign up through our links, at no extra cost to you. This does not affect our editorial independence.

Frequently Asked Questions

Does NAGA accept US clients?

US traders should verify directly with NAGA whether their jurisdiction is supported. Many copy trading platforms offering leveraged forex or CFD products cannot accept US clients due to regulatory restrictions under Dodd-Frank. Check the platform's terms and conditions and verify regulatory status with the relevant authorities.

Can I run an algorithmic bot on NAGA's platform?

NAGA's primary offering is copy trading through its social feed. Whether it supports API-based algorithmic trading should be verified directly with the provider. Most copy trading platforms prioritize social features over algorithmic API access.

What happens if the API connection drops mid-trade on a copy trading platform?

A disconnection during an open position typically means your position is frozen until the connection is restored. The platform does not automatically hedge or close your position. Position sizing that can survive a disconnection is essential.

How does NAGA's profitability affect my trading?

Platform profitability is a positive signal for longevity—it suggests the company is not at immediate risk of insolvency. However, it does not tell you whether the signal providers you are copying are profitable. Evaluate signal provider track records separately.

What is the difference between copy trading and algorithmic trading?

Copy trading replicates the trades of human signal providers. Algorithmic trading uses rules-based code to execute strategies. The key difference is consistency: algorithms follow their code without emotional interference, while human traders are subject to fear, greed, and panic.

How much should I risk per trade on a copy trading platform?

Risk per trade should be based on your total portfolio, not on the signal provider's track record. We suggest risking no more than 1-2 percent of equity per trade, and ensuring that your maximum exposure during a platform disconnection is within that range.

Are there regulatory protections for copy trading clients?

Regulatory protections vary by jurisdiction. Verify the platform's regulatory status with the relevant authority—FCA in the UK, ASIC in Australia, CySEC in Cyprus, or SEC/CFTC in the US. Client fund segregation and compensation schemes depend on the specific regulatory regime.

How do I evaluate a signal provider's track record?

Look for verified track records, not self-reported claims. Check for consistency of returns, maximum drawdown, and the provider's behavior during high-volatility events. Be skeptical of providers with unusually high win rates or returns—they may be taking excessive risk.

Can I stop copy trading cleanly and withdraw my funds?

Withdrawal experience varies by platform. Before committing capital, test the withdrawal process with a small amount. Check for withdrawal fees, processing times, and any restrictions on stopping copy trading relationships. A platform that makes withdrawals difficult is a red flag.

Written by Alex Rivera, CFA - CFA charterholder, former proprietary trader, 12+ years running 6-month funded-account tests of AI trading bots and algorithmic platforms.

Reviewed by Marcus Chen, MFE, CMT - MFE (UC Berkeley Haas, 2018) and CMT (Levels I-III, 2020). Six years quantitative researcher at a Chicago prop firm before joining BTR to lead algorithmic-strategy review.

Read our full Testing Methodology.

Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms