How to Design In-Game Currency for Mobile F2P Games: Soft vs Premium Currency, Faucets and Sinks, Price Ladders, and the 2026 Rules

In-game currency design decides what players earn, how fast, where they spend it, and when they pay. It looks like the work of setting how many coins and gems exist and what they cost, but payer conversion, ARPPU, ARPDAU, and retention all run on top of this structure. The currencies discussed here are closed-loop virtual currencies: they are used only inside the game and cannot be exchanged for cash.

Two things make the topic more pressing in 2026. First, new installs are shrinking. According to a summary of Sensor Tower's 2026 gaming report, mobile game downloads fell 7.2% to 50.41 billion in 2025, while mobile game in-app purchase revenue grew just 1.3% to $81.75 billion. In a market where fewer new players arrive, how existing players move through a game's economy drives revenue. Second, regulators have started looking at currency structures themselves. The EU's Consumer Protection Cooperation (CPC) Network published key principles on in-game virtual currencies in March 2025, and on September 30, 2026, it launched coordinated enforcement against 10 game companies over their virtual currency practices.

Revenue models in general are covered in Game Monetization Strategy: Boosting Engagement and Revenue in 2025, and the mix of IAP and rewarded ads in Balancing IAP and Rewarded Ads Monetization: How Two Revenue Models Work Together. This guide covers the layer underneath, the currency structure, from a marketer's and PM's point of view.

What Types of In-Game Currency Do Mobile F2P Games Use?

Most mobile F2P games are built on two axes, a soft currency earned through play and a premium (hard) currency obtained mainly through purchases, with event or feature currencies added on top for specific purposes.

Unity's game economy design guide, published in August 2026, describes soft currency as "usually earned through play" and taking the form of coins, gold, or energy, and hard currency as "scarce and typically tied to real money." Javier Barnes's taxonomy on Game Developer goes further. He defines hard currency as a high-value currency obtained primarily through IAP and closely tied to monetization, and soft currency as a low-value, general-purpose currency obtained primarily for free. Between them sits medium currency, which can be ground out but is limited in some way, and he separates out feature currencies earned only in specific modes and event currencies that exist only temporarily.

How many currencies to run is a point of disagreement in the industry. In a 2014 PocketGamer.biz panel, most participants agreed that additional currencies let designers reward desired behavior precisely without handing out premium currency. They also warned that a flood of icons and numbers can confuse new players and cause churn, with spikes when a new currency appears. The consensus was that introduction order and tutorials matter more than the number itself. In 2026 there is one more variable: regulation. As covered below, the EU principles list mixing several purchasable currencies, or requiring several exchanges before a purchase, as practices to avoid.

How Do Faucets and Sinks Keep a Currency Economy Stable?

A currency economy stays stable when the ways currency enters (faucets) balance the ways it leaves (sinks); when faucets outpace sinks for long enough, inflation sets in and both progression pacing and purchase motivation break down.

Unity's guide divides an economy into four parts: currencies, faucets, sinks, and converters. Faucets are how players earn currency and how the pace of progression is controlled, sinks are where players spend it, and converters bridge the free and paid sides of the economy. The guide's recommendation is simple: audit every faucet against a matching sink before shipping it, and keep faucet output, including ad rewards, matched to sink capacity at each progression tier.

Inflation usually creeps in quietly. Each update adds a daily reward, an event reward, an ad reward, a comeback reward, but sinks do not grow at the same pace. Long-term players' soft currency balances keep swelling, and the team responds by pricing new content higher. Veteran players stop caring about the currency, new players find everything expensive, and the motivation to pay at a progression wall weakens too. If soft currency can buy almost everything, there is less reason to buy premium currency.

It helps to design sinks by type. Progression sinks such as upgrade costs scale with the growth curve, recurring sinks such as retries and consumables absorb a steady amount every day, and progression-independent sinks such as cosmetics have no ceiling and are good at soaking up surplus currency. At GDC 2014, Vili Lehdonvirta of the Oxford Internet Institute compared six currency sinks and eleven item sinks through basic behavioral economics, explained why players accept some sinks more readily than others, and introduced a spreadsheet method for simulating a sink's impact on the economy.

LiveOps is the most practical way to add sinks on a schedule. Unity's guide recommends adding limited-time sinks, such as seasonal items, when currency starts piling up. Event shops, seasonal cosmetics, and limited upgrade tiers pull accumulated soft currency back out while also giving players a reason to log in. How an event calendar builds retention is covered in LiveOps Strategy: Your Event Calendar Is Your Real Retention System. When evaluating an event from the economy's side, look beyond participation and revenue to how much the median soft currency balance fell from before to after the event.

How Should You Set Exchange Rates and the Premium Currency Price Ladder?

The basics are to anchor the exchange rate to how many minutes of play one unit of premium currency represents, and to design the price ladder's entry price and its upper-tier bonus rates separately.

Unity's guide frames the central economy question as the ratio at which hard currency converts to soft, and warns that getting it wrong in either direction lets one currency cannibalize the other. Its fix is to re-anchor the exchange rate to a fixed unit of time or effort and test it against real player data. For example, if the median player earns 600 coins in a 10-minute session and 100 gems convert to 3,000 coins, then 100 gems are worth 50 minutes of play (illustrative assumption). With that anchor, every new item can be judged consistently by asking how many hours of play it represents. An exchange rate that is too generous makes grinding for coins pointless; one that is too stingy turns the converter into a feature nobody uses.

The price ladder is the set of premium currency bundles and prices in the store. As of October 2026, Apple offers 900 price points from $0.29 to $10,000 in US dollars, so price points themselves are no longer much of a constraint. What remains is how to step up bonus rates. Below is a simplified version of a common shape.

(Illustrative assumption. Actual bonus rates vary by genre and market.)

Marketers should look at two places in this table. One is the bottom row, the entry price for a first purchase. The other is the bonus rate in the upper tiers, which drives repeat purchases and ARPPU among players who have already paid.

First purchase offers (starter packs) usually sit outside this ladder. A 2024 guest post on Deconstructor of Fun by Julian Runge of Northeastern University reported a survey of 54 F2P professionals. 96.3% said they would price a starter pack below $10 and 59.2% below $5, and 61.1% agreed that pricing below willingness to pay is safer. Runge argues that the first purchase price can become an anchor for later spending, so chasing conversion alone ends up sending $2.99 offers to players who would have been interested in a large pack at $19.99 or more. His remedy is experiments that measure long-term effects, plus personalization. That is why a first purchase offer should be judged not only on conversion but on the 90-day ARPPU and repeat purchase rate of the cohort it converts.

The EU principles belong in ladder design too. The CPC Network lists offering currency only in bundles that do not match the value of purchasable items, and denying players the ability to choose the specific amount they buy, as practices to avoid. If a key item in the table above costs 600 gems, the player has to buy two 550-gem bundles or the 1,200-gem bundle and is left with surplus gems. Pricing items to match bundle sizes, or offering the item for direct purchase with real money, reduces the problem.

How Does Currency Design Show Up in Payer Conversion, ARPPU, and ARPDAU?

Most currency design levers show up first in one specific metric, so deciding in advance which metric to watch for each lever makes it far easier to separate cause from effect.

The starting point is that payers are a small minority. In GameAnalytics data covering games from June 2018 to May 2019, the top 15% of games converted 0.8-1.2% of players. The data is old, but the structure it describes still holds: most players experience the economy only through soft currency. Soft currency design therefore mostly determines non-payer retention, while premium currency and the price ladder mostly determine revenue from the small group of payers.

IAP ARPDAU can be broken down into the share of users who paid that day multiplied by the average spend per payer. The same ARPDAU change calls for a different currency lever depending on whether conversion or ARPPU moved. How to read ARPDAU alongside DAU is covered in Revenue = DAU × ARPDAU: The Daily Metric That Separates Real Growth From Install Spikes, and how to read revenue per paying user in our guide to ARPPU in mobile games.

A battle pass is one product that bundles several of these levers. The free track is a soft currency faucet, the paid track is a premium currency destination, and the season's end acts as a limited-time sink. The economics of the pass structure are covered in Battle Pass Economics: How One Season Structure Drives Revenue and Retention at Once.

How Much Currency Should Rewarded Ads and Offerwalls Pay Out Without Cannibalizing IAP?

Rewarded ad and offerwall payouts should be sized by how much that currency substitutes for what players would otherwise buy, and research points to the safest approach being to offer rewards to low-engagement players and scale them back as engagement rises.

Unity's guide calls rewarded ads one of the most important faucets in modern mobile economies, and recommends capping ad-based currency like any other faucet and segmenting ad frequency by spend behavior. Research supports this. A 2022 study by Sheng and colleagues in Manufacturing & Service Operations Management calibrated its model on behavioral data from 5,000 players of a mobile survival game (1,529,577 player-day observations). Among the observation days used in the model, players made a purchase on an average of 4.7% of days and took an incentivized action on 17.4%. The researchers defined how much incentivized actions substitute for purchases as a single variable, the cannibalization parameter, and concluded that it determines how to deploy them. Choosing "always offer" without knowing that parameter could leave performance up to 77% below optimal, and "never offer" up to 85% below. A threshold policy that offers incentivized actions to low-engagement players and removes them past a set engagement level lost at most 6% in the worst case.

There is also evidence that rewards help purchases. A 2017 study in the Journal of Advertising Research found a sampling effect in real game data: users who received a free item as a reward became more likely to buy that item, and existing payers bought more afterward. In an offerwall case study Kongregate published in 2016, payers who engaged with the offerwall had 26% higher ARPPU than payers who did not. Kongregate explained that it set offer payouts to an appropriate value of its hard currency and tiered them by offer complexity.

Put together, sizing payouts becomes a comparison against the price ladder. Suppose a rewarded video pays 10 gems and is capped at three views a day. Over 30 days that is 900 gems, worth about $8.91 at the ladder's undiscounted rate of 101 gems per dollar, more than the $4.99 bundle (illustrative assumption). For a player with no intention of paying, those 900 gems are a sample of what premium currency does; for a player about to make a first purchase, they are a free substitute for it. In practice, teams often pay ad rewards mostly in soft currency or consumables, give premium currency only in small capped amounts, and reduce frequency for the near-payer segment. The cannibalization parameter differs by game, so measure it directly with a holdout: split new users into exposed and unexposed groups and compare IAP ARPDAU and payer conversion over two to four weeks.

The conditions under which offerwalls cannibalize IAP, with benchmarks, are covered in How Much Do Offerwalls Really Lift Retention and Revenue? 2026 Benchmarks and the Conditions That Cannibalize IAP, and the offerwall's role in a hybrid revenue structure in How Offerwalls Fit Into a Hybrid Monetization Strategy.

Rewards that come from outside the game follow the same logic. An external campaign that rewards playtime or specific in-game actions is a way to bring players to a particular point in the economy. Setting the milestone after the player has gone through the economy's first sink loop at least once, for example after completing a first upgrade or joining a first event, means the players who arrive stay with an understanding of what the currency is for. Playio is an Android-based community of five million gamers that analyzes genre preferences, play history, and in-game behavior data to connect players with games that fit their tastes. Pricing is CPI by default, and CPE based on reaching playtime thresholds or completing specific in-game actions is also supported, so campaign goals can be aligned with progression points in a game's currency economy.

You can find more details here. (https://playioadsen.oopy.io/bizdeck)

What Do Apple, Google, and EU Regulators Require of Virtual Currency in 2026?

As of October 2026, the App Store and Google Play require their own billing and odds disclosure for in-game currency sold for money, and the EU is moving into enforcement of principles that add two more demands: show currency prices in real money too, and do not use currency structures to obscure cost.

Apple's App Store Review Guideline 3.1.1 requires in-app purchase for unlocking features or content, explicitly including in-game currencies, and states that credits or in-game currencies purchased via in-app purchase may not expire. The same guideline requires apps offering loot boxes or other mechanisms that provide randomized virtual items for purchase to disclose the odds of receiving each type of item before purchase. The guidelines also state an exception for the United States storefront, where the prohibition on buttons and links directing users to other purchasing mechanisms does not apply. Guideline 3.2.2(x) prohibits forcing users into store-related actions such as downloading other apps, while allowing apps to incentivize in-app actions such as completing a level or watching an ad.

Google Play's payments policy follows a similar structure. In-app products such as virtual currencies, extra lives, and additional playtime must use Google Play's billing system, and in-app virtual currencies must only be used within the app or game title for which they were purchased. Apps selling randomized items must clearly disclose the odds in advance of, and in close and timely proximity to, the purchase. Alternative billing and external links are allowed in eligible regions under program requirements. The fee structure has changed as well: Google applied a new service fee structure in the EEA, UK, and US from June 30, 2026, and in Australia and Japan from September 30, 2026, so the net revenue of a price ladder is worth recalculating by market.

The EU has the biggest effect on design. These are the key principles the CPC Network published on March 21, 2025.

The part designers most often miss is the scope. The principles apply to currencies purchased with real money and explicitly exclude currencies that can only be obtained through gameplay and cannot be bought. Cryptocurrencies and virtual assets as defined under EU anti-money-laundering rules are also excluded, and such tokens are outside the scope of this guide. In other words, keeping event or feature currencies earn-only, never sold for money, takes them out of the principles' scope and out of the mixed-currency problem, and avoids conflict with Apple's no-expiry rule. Once an event currency is sold directly, real-money price display, exchange steps, and expiry all come under review.

The principles are not legally binding, but enforcement has begun. In a joint statement on September 30, 2026, the CPC Network said a high number of game companies had made no substantive changes after the principles and the industry dialogue that followed, and launched coordinated actions asking 10 companies, including King, Supercell, Riot Games, and Ubisoft, to bring their in-game virtual currency practices in line with EU consumer law. The statement added that particular attention is given to variable reward systems such as loot boxes, especially where they are made available to children. Earlier, in September 2024, the European consumer organisation BEUC filed a complaint stating that 21 of the 50 most-played games of 2023 (42%) used premium currencies and calling for in-game prices always to be displayed in real money. Games launched in the EU usually share one currency structure across regions, so treating these standards as the default for the global build is operationally simpler.

Which Currency Design Mistakes Show Up in Which Metrics?

Currency design mistakes are usually discovered only after revenue drops, but most leave symptoms in a specific metric first, so adding the checklist below to a regular review catches them early.

When using the checklist, always split metrics by segment. Cannibalization often shows up as payer conversion holding steady overall while falling only among highly engaged non-payers. For currency balances, the median and the top 10% reveal inflation faster than the average.

Key Takeaways

In-game currency design means building on the two axes of soft and premium currency, adding event and feature currencies where they serve a purpose, and preventing inflation by balancing faucets and sinks. Anchor the exchange rate to how many minutes of play one unit of premium currency is worth, design the price ladder's entry price and upper-tier bonus rates separately, and judge starter packs not only on conversion but on the long-term ARPPU of the cohort they convert. Size rewarded ad and offerwall payouts against the price ladder, and, as the research suggests, combine a threshold approach that offers rewards to low-engagement players and scales them back as engagement rises with holdout measurement. As of October 2026, Apple and Google require their own billing and odds disclosure for paid currencies, and the EU CPC Network is following its principles on real-money price display, avoiding multiple paid currencies and exchange chains, and bundles that match item prices with coordinated enforcement against 10 companies. The fact that earn-only currencies fall outside those principles is a useful rule of thumb when designing event currencies.

For inquiries about Playio's advertising solutions, reach out at: [email protected]

Sources

• GameDev Reports, Sensor Tower: State of Gaming 2026, 2026-03-19 (2025 mobile game IAP revenue $81.75B, +1.3% YoY; downloads 50.41B, -7.2%): https://gamedevreports.substack.com/p/sensor-tower-state-of-gaming-2026

• Unity (Daniel Godley), How to design a game economy that supports IAP, hybrid, and D2C transactions, 2026-08-12 (soft and hard currency definitions; currencies, faucets, sinks, converters; a wrong exchange rate lets one currency cannibalize the other; re-anchor exchange rate to time or effort; audit every faucet against a matching sink before shipping; add limited-time sinks such as seasonal items when currency piles up; rewarded ads as a key faucet, cap ad currency and segment frequency by spend behavior): https://unity.com/resources/game-economy-design-guide

• Game Developer, Types of game currencies in mobile free to play, Javier Barnes, 2021-01-19 (hard, soft, medium, energy, feature, and event currencies; overflowing the economy with hard currency weakens scarcity): https://www.gamedeveloper.com/business/types-of-game-currencies-in-mobile-free-to-play

• PocketGamer.biz, Is the rise of engagement-focused in-game currencies confusing players?, 2014-05-22 (industry panel: control benefits of extra currencies, confusion and churn risk for new players, gradual introduction): https://www.pocketgamer.biz/mobile-mavens/59105/is-the-rise-of-engagement-focused-in-game-currencies-confusing-player/

• GDC Vault, Economic Balancing and Improved Monetization Through Clever Sink Design, Vili Lehdonvirta, GDC 2014 (six currency sinks and eleven item sinks compared through behavioral economics; spreadsheet simulation): https://gdcvault.com/play/1020085/Economic-Balancing-and-Improved-Monetization

• Apple Developer, In-App Purchase (900 price points, US $0.29 to $10,000): https://developer.apple.com/in-app-purchase/

• Deconstructor of Fun, Pricing Starter Packs: When Conversion is King, We May Price Too Low, Julian Runge, 2024-04-08 (survey of 54 F2P professionals: 96.3% below $10, 59.2% below $5, 61.1% agree pricing below willingness to pay is safer; anchoring effect of the first purchase; long-term experiments and personalization): https://www.deconstructoroffun.com/blog/2024/4/8/free-to-play-starter-pack-pricing-when-conversion-is-king-we-may-price-too-low

• WN Hub, How does your mobile game compare in 2019? GameAnalytics provides benchmark stats, 2019-07-01 (June 1, 2018 to May 31, 2019; top 15% of games convert 0.8-1.2%): https://wnhub.io/news/analytics/item-2261

• Sheng, Ryan, Nagarajan, Cheng, Tong, Incentivized Actions in Freemium Games, Manufacturing & Service Operations Management 24(1), 2022 (5,000 players, 1,529,577 player-day observations; purchase days 4.7% vs incentivized-action days 17.4% among model observation days; cannibalization parameter is the key variable; worst-case loss always offer 77.12%, never offer 85.37%, threshold policy 6.00%): https://pubsonline.informs.org/doi/10.1287/msom.2020.0925 (author copy: https://christopher-thomas-ryan.github.io/papers/incented-actions.pdf)

• Lee & Shin, Positive Side Effects of In-App Reward Advertising: Free Items Boost Sales, Journal of Advertising Research 57(3), 2017 (sampling effect of free reward items, more subsequent purchases by existing buyers, moderated by engagement): https://koasas.kaist.ac.kr/handle/10203/226403

• Game Developer, Offerwall Performance Case Study, Jeff Gurian (Kongregate), 2016-06-28 (offerwall-engaged payers +26% ARPPU; offer payouts set to an appropriate value of hard currency and tiered by offer complexity): https://www.gamedeveloper.com/business/offerwall-performance-case-study

• Apple, App Store Review Guidelines (3.1.1 in-game currencies require in-app purchase, purchased credits and currencies may not expire, loot box odds disclosed before purchase; 3.1.1(a) United States storefront exception; 3.2.2(x) no forced store-related actions, incentivizing level completion or ad viewing allowed): https://developer.apple.com/app-store/review/guidelines/

• Google Play Console Help, Payments policy (in-app products including virtual currencies use Google Play billing; virtual currency used only within the purchasing app or game; randomized item odds disclosed before purchase; alternative billing and external links in eligible regions): https://support.google.com/googleplay/android-developer/answer/9858738?hl=en

• Google Play Console Help, Service fees (new fee structure from June 30, 2026 in the EEA, UK, and US, and September 30, 2026 in Australia and Japan): https://support.google.com/googleplay/android-developer/answer/112622

• European Commission, CPC Network Key Principles on In-game Virtual Currencies, 2025-03-21 (principles 1-7; real-money price at the undiscounted rate; avoid mixing currencies and multiple exchanges; avoid bundles mismatching item value; 14-day withdrawal for unused currency; avoid terms allowing unilateral changes to currency value; child protection and spending disabled by default; footnote 2 excludes cryptocurrencies, AMLD virtual currencies, and currencies obtainable solely through gameplay): https://commission.europa.eu/document/download/8af13e88-6540-436c-b137-9853e7fe866a_en?filename=Key%20principles%20on%20in-game%20virtual%20currencies.pdf

• European Commission, Coordinated actions: social media, online games and search engines (background of the principles; industry dialogue did not lead to satisfactory solutions; follow-up coordinated actions): https://commission.europa.eu/topics/consumers/consumer-rights-and-complaints/enforcement-consumer-protection/coordinated-actions/social-media-online-games-and-search-engines_en

• CPC Network, Joint Statement on Stronger Consumer Protection in Video Games across the EU Single Market, 2026-09-30 (high number of companies made no substantive changes; eleven coordinated actions against Activision Blizzard, Crytek, InnoGames, King, Mojang, Plarium, PLR Worldwide Sales, Riot Games, Supercell, and Ubisoft; particular attention to variable reward systems such as loot boxes made available to children): https://commission.europa.eu/document/download/75ff33d4-77a6-4f14-916a-766a8c6c281c_en?filename=CPC%20Network%20Joint%20Statement%20of%20the%20Consumer%20Protection%20Cooperation%20Network%20on%20online%20games.pdf

• BEUC, Consumer groups denounce video games' manipulative spending tactics, 2024-09-12 (21 of the 50 most-played games of 2023 (42%) use premium currencies; call to display in-game prices in real money): https://www.beuc.eu/press-releases/consumer-groups-denounce-video-games-manipulative-spending-tactics

• Own calculations and assumptions: exchange rate example (600 coins per 10 minutes, 100 gems = 3,000 coins); price ladder table (gems per dollar = total gems ÷ price); 500 gems at 101 gems per dollar ≈ $4.95; ad reward 10 gems × 3 views × 30 days = 900 gems ≈ $8.91: illustrative calculations