Abstract: Can this massive trading volume translate into real revenue?

Authors: Changan, Amelia I Biteye Content Team

In the past, discussions about prediction markets focused more on their accuracy, trading volume, and potential to become a new information market. However, when prediction markets are considered a business, the core question changes: what is the profit model for prediction markets?

In the business world, high trading volume does not equate to a platform making money. A trading platform may have a huge following and users may buy and sell frequently, but if most of the transactions cannot be converted into revenue streams, or if activity is maintained purely through subsidies and points, then the trading volume is just a nice-looking statistic, not a healthy source of income.

For prediction markets, the real test of business acumen is not "how many betting platforms are open" or "how popular a certain event is," but rather whether the platform can seamlessly connect these three elements:

The urge to generate real transactions;
Maintain sufficient order book liquidity;
The trading demand (Taker) of actively taking orders is converted into Fees.

This is why the business model of prediction markets is far more than simply "collecting taxes upon opening." On the surface, it's just a series of YES/NO betting games, but what truly underpins the platform's revenue is the underlying trading structure, liquidity mechanism, fee structure, and user behavior.

Especially after leading platform Polymarket began to systematically introduce Taker Fee, the narrative of prediction markets has shifted from "information tool" to "revenue validation".

This article will take a business perspective and deeply analyze the underlying principles of the prediction market:

How do prediction market platforms make money?
Why does the Maker/Taker game structure determine the life or death of a platform?
What are the essential differences in fee design among mainstream platforms, from @Polymarket , Kalshi, @opinionlabsxyz to @predictdotfun ?
Why is the sector with the highest trading volume not necessarily the most profitable?

💡Key conclusion: The market is not selling answers, but rather disagreements.

The closer the price is to 50/50, the greater the market divergence and the stronger the trading impulse, making it easier for the platform to convert transaction fees into revenue. The closer the price is to 0 or 100, the more certain the outcome becomes. Although the information value remains, the corresponding fee weight will decrease significantly.

Therefore, the real business barrier in predicting the market is not turning "events" into betting odds, but turning "disagreements" into transactions, and then steadily converting transactions into revenue.

I. How to make money in the market by predicting opening prices: not by opening prices, but by turning disagreements into feeds.

To break down the cash flow of the forecasting market, we must first clarify its four core revenue drivers. These drivers are intertwined and together form the closed loop from traffic to monetization for the platform.

1️⃣ Transaction fees - Direct source of income

Most prediction markets charge the party that actively trades, known as the Taker. This is because Takers consume liquidity, while Makers provide it.

This means that not all transactions in prediction markets generate revenue. The transactions that truly contribute to the platform's revenue are those that users are willing to actively execute, paying for speed and certainty.

2️⃣ Liquidity - The Cornerstone of Continuous Trading

The hardest part of predicting the market is not the opening price, but making the market have depth.

If there are no orders in a market, and users can't buy or sell, then even if there is discussion in the market, it will be difficult to form an effective price.

Therefore, many platforms will reduce the cost of making a maker, or even provide incentives for makers.

This is not a direct "source of revenue," but it determines whether transaction fees can exist in the long term.

Without liquidity, there is no continuous trading, and transaction fee income naturally cannot be stable.

3️⃣ Information Value - Mindshare

The difference between prediction markets and ordinary trading platforms is that they are not just trading tools, but also generate information.

When a trading platform has sufficient trading volume and liquidity, its price becomes a probabilistic signal. The media will cite it, KOLs will interpret it, traders will observe it, and ordinary users will use it to judge market sentiment.

This portion may not directly translate into transaction fees, but it brings the platform increased attention, user awareness, and external word-of-mouth. In the long run, this information value will, in turn, enhance transaction demand.

4️⃣ User Operations and Discount System - Converting Activity into Revenue

Besides basic transaction fees, different platforms also use discounts, referrals, events, points, and rebates to increase transaction frequency. These measures don't necessarily generate direct revenue, but they do affect the platform's long-term monetization capabilities. For example, Opinion offers user discounts, transaction discounts, and referral discounts; Predict.fun uses a simpler basic fee and discount mechanism; and Polymarket focuses on differentiated rates for different sectors and maker rebates. The essence of discounts and incentives is not simply subsidies, but rather exchanging a portion of profits for user retention, and then gradually converting activity into revenue.

II. Horizontal Comparison of Fee Structures of Mainstream Prediction Market Platforms

Looking at the fee structures of several mainstream prediction markets, the industry's strategic direction is highly convergent: encouraging order placement to provide liquidity and converting active trading into revenue. However, in terms of tactical execution, the major platforms exhibit significant strategic divergence due to their different positioning.

1️⃣ Polymarket: Precise Pricing by Sector

Polymarket's Taker pricing logic perfectly combines "track differentiation" and "discretionary pricing." Its official core formula is:

fee = C × feeRate × p × (1 - p)

Where C is the number of shares traded, p is the transaction price, and feeRate is determined by the market segment.

This mechanism comprises two core variables:

Sector-specific pricing : Based on currently verified fee rates, the fee rate is 0.07 for Crypto, 0.03 for Sports, 0.04 for Politics/Finance/Tech, 0.05 for Culture/Weather, and 0 for some Geopolitics markets. This means Polymarket does not charge a uniform fee for all markets, but rather employs differentiated rates based on the trading frequency, sensitivity, and user willingness to pay for each sector.
Disagreement pricing : perfectly fits the mathematical curve of p × (1 - p). The closer the price is to 50/50 (maximum market disagreement), the higher the commission; the more certain the outcome (closer to 0 or 100), the lower the commission.

https://docs.polymarket.com/trading/fees

2️⃣ Kalshi: Closer to the compliant exchange model

Kalshi's fee structure, within a compliant framework, is closer to that of traditional financial derivatives exchanges, and its standard taker fee formula is also linked to price divergence:

fee = round up(0.07 × C × P × (1 - P))

Where C is the number of contracts, P is the contract price, and the fee is rounded up to the nearest cent. This structure is very similar to Polymarket's C × feeRate × p × (1-p).

Kalshi's fee structure is similar to Polymarket's: its transaction fees are also related to the contract price, with higher fees closer to 50¢ and lower fees closer to 1¢/99¢. Kalshi's fee schedule shows that the taker fee for 100 contracts generally varies between $0.07 and $1.75.

However, a key difference between Kalshi and Polymarket is that Kalshi also charges a Maker fee for some markets, and this fee is only charged when those orders are filled; canceling an order is free. This indicates that Kalshi's fee structure is closer to that of compliant exchanges: it's not simply about making orders permanently free, but rather about setting more complex two-sided fee rules based on different markets.

https://kalshi.com/docs/kalshi-fee-schedule.pdf

3️⃣ Opinion: Emphasis is placed on discounts and user segmentation.

Opinion has introduced a highly complex "multi-dimensional discount system," whose effective rate formula is as follows:

Effective fee rate = topic_rate × price × (1 − price)× (1 − user_discount)× (1 − transaction_discount)× (1 − user_referral_discount)

In other words, Opinion's cost depends not only on the market price and topic_rate, but also on factors such as user discounts, transaction discounts, and referral discounts.

Opinion also sets a minimum order size of $5 and a minimum transaction fee of $0.25 to prevent small transactions from incurring excessively low fees.

This indicates that Opinion's pricing structure is more user-centric:

topic_rate is used to differentiate between different markets.
user_discount is used for user segmentation.

Therefore, compared to Polymarket's "track-differentiated pricing", Opinion is more like turning transaction fees into an operational tool: on the one hand, it guides users to trade, retain and attract new users through a discount system, and on the other hand, it lowers the threshold for placing orders by offering free Maker, thus maintaining market liquidity.

https://docs.opinion.trade/trade-on-opinion.trade/fees

4️⃣ Predict.fun : A minimalist, uniform rate system

Predict.fun has a simpler fee structure, which is suitable for reducing the user's understanding cost.

According to its current publicly available information, its cost calculation formula is:

Raw Fee = Base Fee % × min(Price, 1 − Price) × Shares

The base fee is currently 2%. The actual fee rate will vary depending on the transaction price: below 50%, the fee rate is basically fixed at 2%; above 50%, the closer the price is to 1, the lower the actual fee rate.

In addition, Predict.fun supports fee discounts, which further reduce transaction fees.

The advantage of this design is that it is more intuitive: users do not need to first determine which side the market belongs to, but only need to focus on the transaction price itself to understand the changes in fees.

https://docs.predict.fun/the-basics/predict-fees-and-limits#limits

As can be seen, the common thread among prediction market platforms is that they are all trying to convert proactive trading into revenue.

This also illustrates that there isn't just one path to commercializing prediction markets. Ultimately, they all answer the same question: are users willing to pay for transactions?

III. In-depth analysis of Polymarket: Trading volume does not equal real revenue

Although various platforms offer diverse approaches, Polymarket remains the most suitable platform for observing the true monetization efficiency of prediction markets.

There are two main reasons:

Its pricing strategy is the clearest: from testing the waters with Crypto, to expanding to Sports, and then to charging for almost all categories.
Its data is also more complete: official fee rates and 7D/30D fees can be used to further break down the revenue structure.

So next, we'll take Polymarket as an example to answer a more specific question: Is the sector with the highest trading volume really the most profitable?

3.1 From Free to Paid: Polymarket's Commercialization Timeline

January 2026: Crypto becomes the first paid platform.

Polymarket is returning to its US users by introducing Taker Fees, starting with its Crypto section . Cryptocurrency trading platforms have short settlement cycles, high price volatility, and trading behavior similar to secondary short-term trading. Users prioritize speed of monetization over transaction costs, making it an ideal testing ground for fee-based services.

February 18, 2026: Sports becomes the second paid section.

Following this, on February 18, 2026, the Sports section became the second paid section. Sports betting naturally possesses the characteristics of high frequency and short cycles, providing a continuous trading scenario. Therefore, Sports is a natural continuation of the paid model.

Therefore, Polymarket's decision to charge Crypto and Sports first is essentially a way to validate its revenue model using two more user-acceptable sectors.

March 30, 2026: Fees expanded to more sections

On March 30, 2026, Polymarket expanded its taker fee to include more categories such as Politics, Finance, Economics, Culture, Weather, Tech, Mentions, and Other/General, bringing the total number of fee categories to 10.

After adopting a comprehensive fee structure, Polymarket did not simply charge the same fee across all platforms; instead, it implemented a more granular fee structure. This step can be seen as a key turning point in Polymarket's commercialization, as it began to expand its fee model to a wider market.

The full-fee model has yielded remarkable results. According to the latest data, Polymarket has demonstrated tremendous earning power: 7-day fees reached $9.27 million, and 30-day fees reached $36.3 million . Its 7-day revenue has surged into the top six of all crypto projects, officially entering the ranks of income-generating projects.

3.2 Breakdown of Core Track Single-Type and Price Distribution

To calculate the true revenue of each Polymarket segment as accurately as possible, we estimated the fees for the five main tracks based on Polymarket transaction data from 2021 to February 2026.1

In terms of market price share, the five tracks show significant differences:

The Crypto Market has the highest share at 75%, which perfectly reflects the "rapidly changing" nature of crypto assets, with users preferring to lock in profits and losses directly with market orders. The Weather sector, driven by real-time breaking weather data, also places great emphasis on reaction speed among users.

Secondly, the amount of transaction fees depends heavily on the price range of the order book.

The reason is that transactions entering the fee caliber do not incur the same fees. Polymarket's fees are related to p × (1 - p). The closer the price is to 50/50, the greater the market divergence, and the higher the fee weight; the closer the price is to 0% or 100%, the closer the result is to certainty, and the lower the fee weight.

Data from the five main sectors shows that most transactions were concentrated in the 30-50 range, especially the 40-50 range:

This data shows that Polymarket's main trading activity did not occur in areas where the outcome was close to certain, but rather concentrated in areas where there was still significant disagreement in the market.

3.3 Revenue Calculation: Who is the Cash Cow?

We roughly estimate Polymarket's fee revenue across the five tracks by combining the market trading volume of each track with the corresponding fee rate and weighting it using p × (1-p) for different price ranges. We also consider that after the introduction of fees, some fee-sensitive users will switch from Taker to Limit order placement. Users who engage in end-of-day trading, low-odds arbitrage, or frequent short-term trading will be more cautious in calculating returns.

Therefore, we can make a more conservative assumption based on the original estimate: assume that after the fee is charged, the market price transaction amount of each track will decrease by 20%.

The adjusted formula becomes:

Adjusted estimated fee ≈ Market transaction amount × 80% × feeRate × (1 - p)

Based on the total 7D trading volume and the trading volume share of each track, we estimate the 7D market order trading value of the five major tracks.

Having already calculated the market order value for each track, we will now estimate the fees by combining the fee rate and price range weights for each track. To ensure a more robust calculation, we will use the median of the range as an approximate price:

(Note: Due to statistical caliber, the lag in historical single-type proportions, and dynamic changes in the track, this calculation model aims to restore the contribution ratio of each track. There is a reasonable error between the sum and the total Fees actually settled by the system.)

What does the data tell us?

1️⃣ Crypto is currently the most profitable sector, with an estimated cost of approximately $4.39 million, making it a "cash cow".

This is counterintuitive, because in terms of trading volume, Sport is the largest sector, with 7D trading volume at approximately $401 million, higher than Crypto's $174 million. However, in terms of fees, Crypto ranks first, mainly for two reasons:

Market orders account for a higher percentage: Market orders account for approximately 75%, significantly higher than Sport orders' 60%. Polymarket primarily charges for market orders, so Crypto receives more fees for its transactions.
The feeRate is the highest: feeRate is 0.07, while Sport is only 0.03. Even if both have the same market order size, Crypto will incur significantly higher fees per unit transaction.

2️⃣ Sport is the second largest source of fees, with 7D's estimated fees at approximately $3.31 million, making it the "base of trading volume".

Sport's advantage lies in its large trading volume. Its 7D trading volume is approximately $401 million, ranking first among the five tracks. However, its weakness is also apparent: its fee rate is the lowest at only 0.03.

3️⃣ If Politics and Trump were combined into a single political betting platform, the estimated cost would be around $3.14 million, which is very close to the Sport sector and represents a pulse-like flow funnel.

Political betting platforms are characterized by their strong event-driven nature. Unlike sports betting with its daily stable matches or cryptocurrencies with their continuous price fluctuations, they are prone to concentrated trading during elections, polls, policy changes, and candidate statements. Therefore, while the trading rhythm of political betting platforms may not be stable, their fee contribution is very considerable during periods of high activity.

4️⃣ Weather's 7D estimated cost is around $400,000, the lowest among the five tracks.

Therefore, Polymarket's revenue structure can be simply summarized as follows: Crypto is responsible for platform revenue, Sport is responsible for trading volume, and Politics/Trump is responsible for trending events and customer acquisition for the platform.

IV. Four End-Game Judgments for the Prediction Market Sector Based on Polymarket

Polymarket's successful closed loop offers insights for restructuring the entire prediction market sector:

1️⃣ Complete overhaul of evaluation indicators

In the past, when analyzing prediction markets, people focused on trading volume and trending topics. However, in the commercial era, the metrics for measuring success will completely shift to: real feeds, taker ratio, order book depth, and spread. Trading volume generated purely through short-term trading will be unsustainable under a fee-based system.

2️⃣ Different event types correspond to different income roles.

Future prediction market platforms will not rely on a single betting platform to dominate the market, but will instead move towards a more refined division of labor.

Crypto markets are closer to financial transactions, with rapid price changes and short feedback cycles. Users are more sensitive to the speed of transactions, making it easier to generate high revenue efficiency.
Sports trading is more like a stable flow of funds, with frequent matches, clear results, and a continuous trading environment, making it suitable for contributing to daily trading volume.
Markets like Politics/Trump tend to be more volatile during events, and may not be stable under normal circumstances. However, they tend to see a surge in trading volume during key periods such as elections, polls, and policy changes.
Markets like Weather demonstrate that as long as events are standardized enough and outcomes are clear enough, even if the scale is not large for the time being, there is an opportunity to form its own trading scenarios.

3️⃣ The fee structure will, in turn, force the improvement of betting platform quality.

During the free phase, the platform can open many trading slots; after charging fees, both users and market makers will start to be more careful with their spending, and the fee mechanism will in turn filter the market quality.

A good prediction market not only needs to have interesting topics, but also needs to meet several conditions at the same time:

The results are clear and easy to settle.
Frequent information updates can lead to price changes.
The market divergence is large enough that users have a trading incentive.
With sufficient liquidity, users are willing to actively transact.
The result is not easily manipulated.

4️⃣ The barrier to entry in prediction markets lies in "persistent pricing power".

Opening a YES/NO order book isn't difficult; the challenge lies in ensuring a continuous flow of orders, takers, price updates, and risk-taking. Only when an order book has sufficient depth and trading frequency can its prices be meaningful, allowing the platform to generate revenue.

Therefore, the real barrier to market prediction is not "who can discover hot topics faster", but rather: turning hot topics into tradable markets 👉 ensuring long-term market liquidity 👉 making prices a signal that the outside world is willing to reference.

V. In Conclusion

There are countless projects that can tell grand narratives, but very few can translate those narratives into real financial revenue.

Polymarket was once the most glamorous representative of traffic in the entire sector. Now, having transformed from a "traffic-driven narrative" to a "systematic money-making" approach, it wants to prove one thing to the entire industry:

The ultimate value of prediction markets is not limited to "how accurately they predict the future," but lies in their success in transforming the uncertainties of the real world into a super market where prices can be standardized, transactions can be conducted at high frequencies, and profits can be made continuously.

In the past, prediction markets proved they could generate massive traffic; now, they are proving themselves to be an unparalleled lucrative business.

Disclaimer: All content, data, and opinions contained in this article are for industry exchange and research reference only, and do not constitute any form of investment advice, legal advice, or basis for business decisions.

1. Calculation method: Determine the ratio of market orders and limit orders for each track, then estimate the impact of p × (1 - p) on fees according to different transaction price ranges, and finally calculate the approximate amount of fees contributed by each track by combining the corresponding transaction fee rate.

CFTC Enforcement Director: Combating insider trading and other violations in prediction markets is a priority.

Popular Articles