Imagine you go to an asset manager and they tell you: my fee is 600 dollars a year, regardless of whether I make you money or lose it. If your portfolio drops 5 percent this year, you still pay the 600. If I do nothing and you wait six months for a response, you still pay the 600.

Most investors would never sign that contract. Yet this is exactly the standard in the trading bot market. Subscription, monthly, performance-independent.

This article walks through the four pricing models that exist in the market, calculates their actual costs, and shows why profit-sharing is the fairer model in most scenarios. Including the honest edge cases where subscription is the better choice.

The Four Pricing Models in the Crypto Bot Market

When you decide on a trading bot today, you also decide on a pricing model. There are four variants in the market, and each has a different logic behind it.

Subscription is the most common model. You pay a fixed amount monthly or annually, regardless of how much the bot earns. 3Commas, Cryptohopper, Bitsgap, and Coinrule all work this way. Prices range from 25 to 130 USD per month, tiered by feature scope.

One-Time License means you buy the software once and use it indefinitely. The most well-known example is Gunbot, with lifetime licenses between 199 and 500 USD. Sounds appealing, but has its own pitfalls.

Free Tier is usually a hook. Pionex offers its bots for free, but earns money through the trading fees your bots generate on its own exchange. Open-source frameworks like Freqtrade are actually free, but you pay with your time on setup and maintenance.

Profit-Sharing is the youngest and rarest model. You pay nothing upfront. Instead, the provider keeps a percentage of your realized profits, typically between 20 and 30 percent. If the bot earns nothing, you pay nothing. unCoded is currently the only commercial product in the spot trading space that operates on this model.

The question is not which model is theoretically better. The question is which model leads to which performance and what actually remains of your profit at the end.

Subscription: Who Really Benefits?

The math behind subscriptions is simple and unforgiving. You pay a fixed amount, no matter what.

Take the 3Commas Pro plan as an example. It costs 59 USD per month. Over twelve months, that adds up to 708 USD. For you to be in the green at the end of the year, your bot needs to earn at least 708 USD, plus your exchange's trading fees, plus any costs for tax tools.

With 10,000 USD of capital, that means your bot needs to deliver at least 7.1 percent return just for the subscription model to break even. Anything below that means you're paying money to the provider without earning anything yourself.

Sounds doable? Let's look at reality.

Weak year (5 percent performance, 500 USD profit): Your bot generates 500 USD. You pay 708 USD subscription plus around 80 USD in trading fees. Net loss of 288 USD.

Standard year (10 percent performance, 1,000 USD profit): Your bot earns 1,000 USD. Subscription and fees come to 788 USD. You keep 212 USD of your 1,000 USD profit. That's 21 percent.

Good year (25 percent performance, 2,500 USD profit): Your bot earns 2,500 USD. Fixed costs stay at 788 USD. You keep 1,712 USD. That's 68 percent.

So subscription rewards good performance and punishes bad performance. But the punishment is asymmetric. In a weak year you lose money, even though your bot was profitable. That's structurally problematic.

Who profits from this model? The provider. Their revenue is predictable, independent of bot performance. If the bot runs poorly during a market phase, that doesn't affect the provider. They keep earning, you keep paying.

Lifetime License: The Middle Ground Trap

A lifetime license looks attractive at first glance. Instead of paying monthly, you spend 199 or 500 USD once and use the software indefinitely.

The math looks good. With a lifetime price of 199 USD and an alternative subscription price of 59 USD per month, you reach break-even after just under four months. Everything after that is saved.

But there are three catches.

First, the update path. Lifetime with most providers means: lifetime of the current major version. When the provider releases version 4, your lifetime license for version 3 doesn't cover it. You can keep using what you have, but updates cost extra. Sometimes nearly the full price again.

Second, the effort involved. Lifetime licenses are mostly issued for self-hosted software that you have to run on your own server. That's not inherently bad (more on this in our blog post on non-custodial trading bots), but it means additional setup effort and operational responsibility.

Third, the performance problem isn't solved. A lifetime license is cheaper than a subscription, but it has the same structural weakness: you pay regardless of whether the bot makes money. In a bad year, the one-time investment is bearable, but it still sits in your books. And if the bot stops working after three months (because market conditions changed or the provider discontinued support), the money was spent for nothing.

Lifetime models are better than monthly subscriptions, but they're not a game-changer. They just push the problem into the future.

Free Tier: The Hook

Free trading bots sound like the best deal of all. But anyone working for free earns their money somewhere else. The only question is where.

Pionex offers 16 bots for free, but at the same time operates the exchange where these bots trade. The revenue model runs through trading fees. Pionex takes 0.05 percent per trade, which is lower than Binance (0.1 percent), but with an active bot the amounts add up over the year. Something rarely mentioned in Pionex marketing materials: spreads between buy and sell prices on smaller exchanges are often worse than on Binance or other large platforms. A bot that gets a less favorable spread on every trade systematically makes worse trades. These hidden costs don't appear in any invoice, but they reduce your real return. And because the spread benefits the platform directly, Pionex has a structural interest in keeping that gap from getting too tight. The actual catch, though, is custody: your capital sits on Pionex, not on an exchange of your choice. You give up custody control to save on subscription costs.

3Commas Free Tier also exists, but it's so heavily restricted that serious use is barely possible. Maximum three active bots, no advanced strategies, no backtest. The free tier is a funnel to the Pro plan, not a complete product.

Open-source frameworks like Freqtrade or Hummingbot are actually free. There's no hidden revenue model because no company stands behind them. But the price you pay is time. Setup, configuration, maintenance, debugging, updates, and strategy development all fall on you. And when something doesn't work, there's no support to turn to. You depend on community forums, GitHub issues, and Discord servers, where someone might help you if they have the time and inclination. A commercial provider with a dedicated support team is structurally different here: when you have a problem, there's someone who is paid to solve it.

So free tier is either a hook with hidden custody risk, a restricted funnel to a paid model, or a genuinely free tool that requires massive personal effort. There is no fourth variant.

Profit-Sharing: The Aligned Incentive Model

Profit-sharing works differently from all other models. The provider only earns when you earn.

The mechanics are simple. The bot tracks every realized profit. At the end of a billing period (typically a month), a percentage of that profit is transferred to the provider. With unCoded, that's 30 percent in the standard variant, reduced to 20 percent with the community bonus.

If your bot earns no profit in a given month, you pay nothing. If it loses money, you also pay nothing. A share is only owed when realized profits actually exist.

What does "realized" mean? This is important to understand. A profit is realized when a position has been closed and the result is in the green. An open position with a paper gain is not a realized profit. Only actually completed trades count. This protects you from paying on paper gains that might disappear later.

What happens in a loss scenario? This is where profit-sharing models in the market differ significantly. unCoded solves it in its own way: the bot only sells a position when it's in the profit zone. That means every closed trade is profitable by definition. Paper losses on open positions are possible (that's part of any trading strategy), but they don't get realized. As a result, users don't experience "losing months" in the classical sense, where the balance dips into the red.

Structurally, the profit-sharing model creates an alignment of interests that doesn't exist in any other pricing model: the provider only earns when the user earns. With subscription bots, the bot's performance is economically irrelevant to the provider. Revenue flows regardless. With profit-sharing, the provider has a direct financial interest in the bot actually working well. When an update degrades performance, it's not just user satisfaction that drops, the provider's own revenue drops too. That's a built-in quality pressure that no subscription model has.

Total Cost of Ownership: The Honest Calculation

Theory is one thing, concrete numbers are another. Here are the calculated scenarios for a typical setup: 10,000 USD capital, 12 months runtime, average trade frequency. The performance values (5%, 10%, 25% per year) are deliberately conservative, chosen to create a comparable baseline. They are not return promises and do not refer to any specific bot. They simply show how different pricing models behave at different success levels.

Scenario 1: Weak year (5 percent performance, 500 USD gross profit)

3Commas Pro (Subscription): 708 USD subscription + 80 USD fees = 788 USD total costs. Net loss of 288 USD.
Cryptohopper Adventurer: 588 USD subscription + 80 USD fees = 668 USD total costs. Net loss of 168 USD.
Bitsgap Advanced: 768 USD subscription + 80 USD fees = 848 USD total costs. Net loss of 348 USD.
Pionex (Free): 50 USD fees only. Net profit of 450 USD.
Profit-Sharing 30% (unCoded): 150 USD profit share + 80 USD fees + 84 USD VPS = 314 USD total costs. Net profit of 186 USD.

In a weak year, every subscription bot loses money. You earned 500 USD and paid over 600 USD in costs. That's a net loss, even though the bot was actually profitable. Pionex looks best here because no subscription is involved, but your capital sits on an exchange you didn't choose. Profit-sharing stays in the green because the share scales with performance.

Scenario 2: Standard year (10 percent performance, 1,000 USD gross profit)

3Commas Pro: 788 USD total costs. Net profit of 212 USD.
Cryptohopper Adventurer: 668 USD total costs. Net profit of 332 USD.
Bitsgap Advanced: 848 USD total costs. Net profit of 152 USD.
Pionex: 50 USD total costs. Net profit of 950 USD.
Profit-Sharing 30%: 300 USD profit share + 80 USD fees + 84 USD VPS = 464 USD total costs. Net profit of 536 USD.

With average performance, varying amounts remain from 1,000 USD gross profit after all costs. Subscriptions eat up 47 to 85 percent of the profit. Profit-sharing takes 46 percent, including VPS and trading fees. Pionex is nominally best, but the hidden custody risk can't be quantified in dollars.

Scenario 3: Good year (25 percent performance, 2,500 USD gross profit)

3Commas Pro: 788 USD total costs. Net profit of 1,712 USD.
Cryptohopper Adventurer: 668 USD total costs. Net profit of 1,832 USD.
Bitsgap Advanced: 848 USD total costs. Net profit of 1,652 USD.
Pionex: 50 USD total costs. Net profit of 2,450 USD.
Profit-Sharing 30%: 750 USD profit share + 80 USD fees + 84 USD VPS = 914 USD total costs. Net profit of 1,586 USD.

In a strong year the relationships flip. Subscriptions are now in the lead because their costs stay fixed and the gross profit is significantly higher. Profit-sharing becomes relatively more expensive, because 30 percent of a high profit is nominally more than a fixed subscription amount. With the community bonus (20 percent instead of 30), unCoded comes to 580 USD plus fees plus VPS, which is 744 USD, putting it at a similar level to the more expensive subscriptions.

What the numbers show: Profit-sharing is structurally superior in weak and average years. In very good years, subscriptions become competitive, but they don't pull away dramatically. Across multiple years where good and bad phases alternate, profit-sharing is almost always the calmer route.

When Subscription Beats Profit-Sharing

Honesty belongs in the analysis. There are scenarios where subscription models are the more rational choice.

When you expect very high profits and actually achieve them. With consistently above-average performance, profit-sharing eventually becomes more expensive than any subscription. Anyone running a bot that reliably makes 50 percent or more per year is nominally better off with a fixed subscription. Important to understand: even in this case, profit-sharing is not a losing proposition, just more expensive than the alternative. You still earn a lot of money, you just give up a larger absolute amount. Anyone who wants to switch under these conditions can do so at any time.

When you prefer predictable costs. Some investors and especially businesses want to know what software costs per month, without the number fluctuating. Profit-sharing is variable, which is a disadvantage for accounting. Anyone who values fixed line items in their budget is better served by subscription models.

When you deploy very large capital. For amounts beyond 100,000 USD, profit-sharing quickly turns into significant absolute sums. 30 percent of 10,000 USD profit is 3,000 USD. Subscription bots stay constant in this range at 600 to 1,500 USD per year. Here it makes sense to negotiate individual conditions (many profit-sharing providers offer volume discounts) or switch to a subscription.

When you use the bot rarely or only short-term. Anyone who only wants to test a bot for one or two months pays less with a subscription than with profit-sharing, assuming high performance during that period.

In all other cases, meaning normal capital sizes, average performance expectations, and long-term use, profit-sharing is the fairer model.

Red Flags in Pricing Models

Whichever model you choose, there are warning signs that are problematic regardless of pricing.

Auto-renewal without a clear cancellation policy. Some providers automatically extend annual subscriptions and deliberately make cancellation complicated. Check before purchase how and by when you can cancel.

Hidden fees on trade volume. Some bots charge additional fees on trading volume or realized profits on top of the subscription. That's a kind of profit-sharing layered onto a subscription. Paying twice is not acceptable.

Tier jumps with absurd price increases. A bot that jumps from 29 USD in the basic plan to 299 USD in the next tier has a questionable pricing logic. Usually the important features are only available in the most expensive tier.

Lock-in through missing export functions. If you can't export your configurations, trade histories, and strategy settings, you're tied to the provider. Switching becomes a complete reinstallation. That's a hidden switching cost.

Performance guarantees. Any provider promising "guaranteed profits" or assuring specific returns has a legal problem. Trading performance can't be guaranteed. Anyone claiming otherwise is either lying or operating in a legal gray area.

Profit-sharing without transparent calculation. There are bad actors in profit-sharing too. If the provider doesn't clearly disclose how exactly the profit is calculated (which trades count, how losses are offset, what currency is used for billing), you have no control. Demand a sample calculation before signing.

The unCoded Model in Detail

Since profit-sharing is unCoded's central differentiator, here are the concrete numbers without marketing wrapping.

Standard share: 30 percent on realized profits. If your bot earns 1,000 USD in a month, 300 USD flows to unCoded. If it earns nothing, nothing flows.

With community bonus: 20 percent instead of 30 percent.

Volume discount: For very large trading volumes, individual conditions can be negotiated. This is relevant for institutional investors and family offices.

Test license: 25 USD for a time-limited test phase with real trades. This lets you try out the system before committing to the profit-sharing model.

What's not included in the profit share: Trading fees on Binance (standard 0.1 percent per trade) go directly to Binance, not to unCoded. The VPS the software runs on is your own infrastructure and costs between 5 and 15 EUR per month depending on the provider. Both items are already included in the TCO calculation above.

How billing works: The share is calculated automatically and displayed transparently in the unCoded dashboard. You can see at any time which trades contribute to the calculation, how high the current profit share is, and when the next billing happens. Nothing runs in the background, everything is traceable.

How the system handles losses: unCoded only sells a position when it's in the profit zone. There are no forced sales in the loss zone. Paper losses on open positions are possible, but they don't get realized. As a result, users don't experience months where closed trades sum up negatively and distort the next profit-share calculation.

FAQ

What happens when the bot makes a loss? With unCoded, this situation doesn't structurally arise. The bot only sells a position when it's in the profit zone, so closed trades are profitable by definition. Paper losses on open positions can occur, that's part of any trading strategy, but they don't affect the profit-share calculation because only realized (i.e. closed) trades count. With other profit-sharing providers that also sell in the loss zone, there are sometimes loss-offset mechanisms. With unCoded, those aren't necessary.

Are 30 percent profit share fair? Compared to subscriptions in most performance scenarios, yes. In a weak year, profit-sharing is clearly superior. In a standard year, roughly equivalent or better. In a very good year, slightly more expensive. The community bonus reduces the share to 20 percent.

Are there hidden fees? With a serious profit-sharing model, no. With unCoded, the only costs beyond the profit share are your exchange's trading fees (which go directly to Binance) and the VPS costs (which go directly to your hosting provider). There are no setup fees, no account fees, and no hidden tier jumps.

Can I cancel at any time? Yes. Since there's no subscription model, there are no cancellation periods. You can stop the bot at any time and shut down the system. What's been billed is billed, new profit shares only arise from new realized profits.

What about open-source bots, are they really free? Yes, in terms of software license costs. But they're not free if you count your time as a cost factor. Setup, configuration, maintenance, updates, and strategy development add up to several hours per week. Anyone with the technical background can invest that time well. Anyone without often fails at installation or ends up with a bot that doesn't run reliably.

Why don't more providers offer profit-sharing? Because it's technically and administratively more demanding. Subscription is simple: a fixed amount per month, predictable revenue, easy bookkeeping. Profit-sharing requires reliable performance tracking infrastructure, transparent billing, a mechanism for handling losses, and in many countries different tax treatment. The business model is also less predictable for providers: in weak market phases, revenue drops, which means financial risk. Subscription shifts that risk entirely onto the users. Profit-sharing shares it.

What's the difference between profit-sharing and a performance fee? Essentially the same thing. "Performance fee" is the term from classical asset management (hedge funds, private equity). "Profit share" is the term that has become established in the crypto space. Both describe success-dependent compensation that only applies when actual profits exist.

Is profit-sharing also worth it for small capital? Especially for small capital, it's the only sensible model. With 1,000 USD of capital and a 10 percent return (100 USD profit), you pay 30 USD profit share. A subscription of 59 USD per month would consume the same capital within two months, long before any meaningful profits could even arise. Profit-sharing makes entry into automated trading viable for small accounts in the first place.

What happens to my profit if the provider goes out of business? With a self-hosted profit-sharing model, nothing. Your capital is on your Binance account, the bot runs on your server. If the provider disappears, the bot keeps running until you manually shut it down. There are no more updates, but your money isn't affected. This is one of the structural advantages of non-custodial self-hosted setups. More on this in our blog post on non-custodial trading bots.

Can I switch from a subscription model to profit-sharing? If you're currently using a subscription bot and considering a switch: yes, this is straightforward. You cancel the existing subscription, export your trade history and strategy configuration (if possible), and set up the new bot. With self-hosted solutions, you also need a VPS. The switch itself takes one to three hours, depending on your technical background.

Why Profit-Sharing Beats Subscription: Trading Bot Pricing Models 2026 Explained