Pricing for Platforms: How to Charge When You’re the Ecosystem

Platform businesses operate multi-sided ecosystems connecting two or more distinct user groups (e.g. buyers and sellers, developers and end-users). Unlike traditional products, platforms must carefully decide which side(s) to charge and how, because pricing one side affects participation and value on the other side. The goal is to maximize network effects and growth while capturing value. This makes platform pricing a complex strategic challenge: set prices too high on one side and you may drive away users, but charge nothing and you may leave value uncaptured. In this research, we explore principles and models for pricing in platform ecosystems, with examples spanning SaaS marketplaces, developer platforms, B2B ecosystems, consumer app stores, and more (primarily in software and digital domains). We examine common multi-sided pricing models and survey how leading platform companies – from Apple and Shopify to Salesforce, Stripe, Amazon, Uber, and others – structure their fees.

Key Principles of Platform Pricing

Network Effects and Externalities: Successful platforms create value by enabling direct interactions between user groups. Pricing has to account for cross-side network effects: charging one group can impact the growth and engagement of the other. For example, if a marketplace raises fees on sellers, they may list fewer products or raise prices, reducing selection or increasing costs for buyers – which in turn could push buyers away. This feedback loop means pricing decisions on one side have indirect effects on the whole ecosystem’s health. Platform pricing must therefore be designed to enhance the overall value proposition for all sides, not just maximize short-term revenue.

“Subsidy Side” vs “Money Side”: Often, platforms do not charge all sides equally – pricing is frequently skewed to subsidize the side that is harder to attract or less willing to pay. A classic strategy is to monetize the side that derives the most value from the platform, while heavily discounting or even subsidizing the other side. For instance, OpenTable initially charged restaurants (who gained reservations) while actually rewarding diners with loyalty points (negative price) to encourage adoption. Similarly, in video game platforms, console makers often sell devices at a loss to gamers, recouping revenue by charging game developers royalties – because players are price-sensitive and developers gain access to a large user base. The rule of thumb is to charge the side with higher willingness to pay or stronger dependence, and keep prices low (or zero) for the side that increases the network’s value by joining.

Avoiding Friction to Build Critical Mass: In early stages, platforms benefit from minimizing “pricing friction” that slows user acquisition. Often this means offering free or very cheap access to one or all sides until the network reaches critical mass. Many platforms start with a freemium or zero-fee model to encourage rapid growth. For example, LinkedIn grew by keeping the service free for users and only later monetizing via premium subscriptions and recruiters/advertisers. Facebook famously refused to charge users (to avoid hampering growth) and instead monetized via advertising once the user base was massive. The key is that pricing should not create a barrier to joining the platform; instead it can be introduced or increased after the network effects take hold.

Balancing Cross-Side Value: An optimal platform pricing structure balances the needs of both supply and demand. If one side of the market cannot easily “multihome” (participate in multiple platforms) but the other side can, the platform can charge more to the captive side. For example, game developers often release games on multiple consoles, but many consumers will only buy one console – thus console platforms charge gamers for hardware and subscriptions, while courting developers with incentives. In general, the side that receives more value from the interaction often pays more. However, platforms must be cautious: just because one side gains significant value doesn’t always mean that value can be monetized without consequence. If charging a certain side would shrink the network too much, alternatives like subsidies or third-party monetization (ads, data, etc.) might be wiser.

Preventing Disintermediation: Platforms only capture value if transactions actually flow through the platform. If users can take transactions off-platform to avoid fees, a high direct fee may fail. In scenarios where the core transaction is hard to monitor or enforce on-platform, many platforms use upfront or subscription charges instead of per-transaction fees. For example, dating platforms cannot easily track whether two users meet in person (the “transaction”), so they often charge a subscription or membership fee to use the service. This captures value upfront and encourages engagement (unlimited interactions once paid) without relying on an honor system for each “date”. Similarly, lead-generation services (connecting contractors with clients, etc.) may charge per lead or a listing fee rather than a cut of an untracked final deal. The general principle is to align the pricing model with the point in the user journey where the platform has control or can demonstrate value, thus minimizing off-platform “leakage” of value.

Common Platform Pricing Models

Platforms employ a variety of monetization models, often in combination. Here are the major pricing models used in multi-sided platforms:

• Transaction Fees / Commission: Charging a percentage or flat fee per transaction facilitated by the platform. This is one of the most common models for marketplaces and app platforms. It aligns monetization with successful matches. Examples: eBay and Amazon charge sellers a referral fee on each sale (typically in the ~8%–15% range depending on category). Apple’s App Store and Google Play take a commission on app sales and in-app purchases (standard 30%, with reduced 15% rates for small developers). Ride-sharing and delivery platforms (Uber, Lyft, DoorDash, etc.) also take a cut of each fare or order (often around 20–30%). Transaction fees have the advantage of “no cure, no pay”, that is users only pay when they get value (a completed transaction). This encourages participation initially, but high take-rates can become contentious as ecosystems mature (leading to pressure for lower commissions).
• Subscription or Membership Fees: Charging a recurring fee for access to the platform or premium features. This model is common when the platform provides ongoing service or when transactions are difficult to meter. Examples: Many dating apps and B2B platforms use subscription models so that users pay for access and unlimited usage. Professional networks like LinkedIn offer premium subscriptions (for recruiters or power-users) that grant advanced search and messaging capabilities – essentially monetizing heavy usage. Amazon’s “Professional” seller plan charges merchants a monthly subscription ($39.99) for advanced selling tools and lower per-item fees. Membership fees can also signal quality: by charging the supply side a joining fee, platforms like early Alibaba ensured only serious sellers participated, and even used those fees to fund buyer referral incentives. This model guarantees some revenue and commitment, but it introduces up-front friction, so it’s often combined with free tiers or trials.
• Freemium Model (Free + Premium Upgrades): Offering basic platform access for free to all users and charging for enhanced features or capabilities. Freemium reduces initial friction to zero, fueling growth, while still allowing monetization of the most engaged users. Examples: Many SaaS platforms and developer ecosystems adopt freemium: e.g. API platforms might offer a free tier for developers (limited usage) and charge for higher volumes or enterprise features. LinkedIn is free for general networking but charges for Sales Navigator or recruiting tools. Streaming and content platforms (Spotify, YouTube) let users participate for free with limited features or ads, while offering an ad-free or feature-rich experience at a premium subscription. Advertising support often complements freemium – effectively monetizing free users indirectly. As noted, ad-supported freemium has been a key to scaling user bases (e.g. Facebook’s entire consumer side is free, with revenue coming from advertisers). The challenge is balancing what to give away vs. hold back for premium without harming the network effects.
• Listing Fees or Lead Fees: Charging sellers or providers to list their offerings, or to receive leads, regardless of outcome. This model is used when quality control or initial matching is the main value, rather than guaranteeing a transaction. Examples: Historically, eBay charged listing insertion fees (a small fee to post an auction) to prevent an overload of low-quality listings. Many job boards and freelance marketplaces charge employers to post job listings or projects (upfront), rather than taking a cut of the employee’s salary. Home services platforms like RatedPeople charge contractors a lead fee for customer inquiries – since the actual job may be agreed offline, the platform ensures it gets paid for the introduction. Listing fees can discourage spam and offset platform costs, but they may also deter participation if too high; thus some platforms later shift to transaction-based fees as technology allows tracking more interactions.
• Revenue Share with Complementors: For developer-centric platforms and app ecosystems, a typical model is a revenue share on sales generated by third-party complementors. This is essentially a commission, but worth noting in context. Examples: App marketplaces often standardize a revenue split (e.g. 70/30 or 85/15). Apple’s App Store takes 30% of app sales and in-app purchases (15% for small businesses under $1M revenue). Salesforce AppExchange takes a 15% cut of ISV partners’ subscription revenues for apps built on Salesforce (or 25% if the app is fully “OEM” white-labeled). Shopify’s App Store similarly charged 20% historically, and now takes 15% of app revenue (after an initial $1M exemption). This model aligns the platform’s income with the success of its ecosystem partners, and it encourages the platform provider to invest in third-party success (since it directly shares in upside). However, the level of the revenue share can become a competitive factor – for instance, to attract developers, some platforms have lowered their cut (Shopify dropped to 0% on the first $1M to woo developers during 2020–2021; Epic Games Store takes only 12% vs. the 30% typical on Steam).
• Payment Processing and Financial Fees: Some platforms monetize by integrating financial services and taking a margin on payments. This often overlaps with other models but deserves mention for financial platforms or any ecosystem handling money flow. Examples: Stripe, a developer payment platform, doesn’t charge a traditional subscription or license – instead it earns via a per-transaction fee (commonly 2.9% + $0.30 for online card payments). Essentially, Stripe provides the platform “for free” and monetizes each payment processed. Similarly, PayPal and other payment networks charge transaction fees to merchants. In marketplaces like Shopify or Etsy, payment processing is a revenue stream: Shopify charges ~2.9% + 30¢ on transactions via its Shopify Payments, and if a merchant opts for an external payment gateway, Shopify adds an extra 0.5–2% fee, effectively monetizing payments. Credit card networks are multi-sided platforms that charge merchant fees (interchange and network fees) while cardholders often pay nothing or even get rewards – the platform makes money by taking a slice of transaction value in exchange for facilitating the exchange. In summary, enabling and skimming the monetary flows in the ecosystem is a powerful model, especially when users are price-sensitive to upfront fees but less so to a small cut of each transaction.
• Advertising and Data Monetization: As noted, some of the world’s largest platforms monetize a third party. By giving free access to users, platforms like search engines or social networks gather large audiences, then charge advertisers for exposure or insight. Examples: Google Search and Facebook/Instagram operate on an ad-supported model – users and content providers pay nothing; advertisers pay for targeted access. This can be seen as a two-sided pricing strategy: one side (users) has a price of $0, the other side (advertisers) effectively pay the platform’s bills. This model works when having a high volume of users is the key to value (thus you maximize users by keeping it free and frictionless, then monetize their attention). It’s worth noting that even non-ad platforms sometimes add advertising as a supplemental revenue stream (e.g. Amazon charges sellers commission but also makes money by selling ad placements in the marketplace; app stores might feature sponsored apps). Data monetization can be a variant – e.g. platforms selling anonymized insights or using data for targeted marketing – though directly selling user data is sensitive and less common in consumer platforms (due to privacy). Nonetheless, advertising revenue is essentially monetizing the externality of having a big network, rather than charging the core users for basic usage.
• Value-Added Services and Upsells: Platforms can also charge for premium services, tools, or enhanced visibility. These can be thought of as ancillary pricing strategies on top of core models. Examples: Many marketplaces offer promoted listings or advertising to sellers for a fee (for instance, Etsy and eBay allow sellers to pay extra for better visibility). Enterprise platforms might offer premium analytics, API access, or support for a higher price. Cloud platforms (AWS, Azure) have marketplaces where the platform not only takes a listing fee but may also upsell vendors on co-marketing packages or enhanced placement. Some platforms charge for certification or trust signals (for example, Airbnb hosts can pay for professional photography; App stores charge for developer programs or security reviews). While not a primary revenue model on their own, these tactics allow a platform to monetize power users or businesses beyond the standard fees – effectively tiering the ecosystem by willingness to pay for additional benefits.

Most platforms employ a mix of these models. For instance, a marketplace might have a listing fee and a transaction fee and offer premium listings for an ad fee. The right combination depends on the platform’s specific dynamics: the goal is to maximize participation and transactions while capturing a fair share of the value created without stifling growth or encouraging defection.

Multi-Sided Pricing Strategies and Considerations

Designing platform pricing is not just picking a model in isolation – it requires an integrated strategy across the different sides of the market:

1. Who (and How) to Charge – One Side or Both?
Platforms must decide whether to monetize one side heavily while subsidizing the other, or to charge both sides to some degree. Many successful platforms start by charging only one side (the one with clearer ROI or less elasticity). For example, eBay historically charged sellers fees but made buying free – sellers were willing to pay for access to many buyers, and keeping it free for buyers encouraged market growth. OpenTable charges restaurants per reservation and subscription fees, while diners not only use it free but earn points – the diners attract the restaurants, so the diners are subsidized. Facebook and most social media charge only the advertiser side, never the end-user. In contrast, some platforms do charge both sides, though often asymmetrically. Airbnb is a prominent example, with a traditional “split fee” model: the guest pays roughly 14% of the booking as a service fee, and the host pays ~3% from their payout. This split approach shares the burden, though not equally. Another example is Upwork, a freelance marketplace: clients (buyers) pay a small processing fee (~3%) and can opt for monthly plans for added benefits, while freelancers (sellers) pay a sliding commission (5–20% of earnings) on each contract. Charging both sides can maximize revenue, but it must be balanced so neither side’s participation is unduly deterred. In Airbnb’s case, initially the platform charged the demand side more (guests) because early on it was harder to attract hosts – hosts kept nearly all their asking price. As the platform grew and supply became plentiful, Airbnb increased host fees for professional hosts (even moving toward a host-only 15% fee option in some cases). This highlights that pricing strategies may evolve over a platform’s lifecycle – who you charge, and how much, can shift as relative network power shifts.

2. Subsidies and Incentives:
It is common for one side of the platform to be subsidized or even receive incentives to participate, funded by monetizing the other side. Sometimes this goes as far as negative pricing (paying users) to overcome a cold start. For instance, PayPal famously paid bonuses to new users and referrers in its early days to build scale on its payments platform. Rideshare companies spent heavily on driver and rider incentives (bonuses, discounts) in their growth phases – effectively charging far less than cost on one or both sides, funded by investor capital, to build the network. While not sustainable long-term, these subsidies are an extreme form of price strategy to ignite network effects. Even in steady state, some platforms maintain ongoing subsidies: OpenTable awarding diner points (which cost the company money) is a permanent incentive to keep diners using the platform for reservations. In more subtle ways, freemium models can be viewed as subsidies – free users are subsidized by paid users or advertisers. The key consideration is that the subsidy side usually has high value to the paying side (i.e. bringing them in makes the whole platform more valuable). Platforms carefully calculate customer acquisition cost vs. lifetime value on each side to decide if subsidies make sense. Many platform business models rely on cross-side subsidies until a critical mass is reached and monetization can ramp up.

3. Preventing Disintermediation and Leakage:
As mentioned earlier, if a platform’s pricing is too aggressive on transactions, users may find ways to bypass it. Platforms must align their pricing model with where they add undeniable value and can enforce participation. This is why some platforms choose upfront fees (listing or subscription) – once paid, the user can use the service freely without incentive to circumvent each transaction. Others provide so much convenience or safety in the transaction process that users remain willing to pay fees (e.g. Airbnb provides payment escrow, insurance, and a large audience – hosts and guests generally complete the booking on-platform despite knowing the fee, because of trust and convenience). If disintermediation is a risk, platforms might lower fees to an acceptable “tolerated” level. For example, if a freelance platform charges 50% commission, clients and freelancers would likely move off-platform after the first contact; but at ~10-20%, many stay because the platform provides payment protection, search convenience, and dispute resolution worth that fee. In cases like dating apps or classifieds where the platform can’t easily meter the exchange, subscription models or advertising models dominate to ensure the platform still profits. Another tactic to prevent off-platform deals is adding value around the transaction – for instance, Upwork’s escrow and record-keeping, or Fiverr’s integrated workflow – making it easier to just stay and pay the fee. Pricing strategy thus must consider human behavior: set it so that the path of least resistance is to use (and pay) the platform, rather than circumvent it.

4. Dynamic and Tiered Pricing Strategies:
Platforms often employ tiered pricing or dynamic fees to fine-tune the balance. Tiered revenue share is one approach: for example, Upwork lowers the freelancer’s commission rate as the lifetime billings with a client increase (20% for first $500, 10% up to $10k, 5% beyond), encouraging long-term projects to remain on the platform. App stores have introduced tiered commissions – Apple and Google cut their fee from 30% to 15% for small developers under $1M revenue, which keeps indie developers happy while the big fish still pay standard rates. Shopify’s app store implemented a 0% fee on the first $1M annually for developers and 15% after that, though in 2025 they adjusted it to a one-time lifetime $1M exemption. These tiered schemes recognize that one size doesn’t fit all – smaller participants are more price-sensitive and need encouragement, whereas larger participants (who derive more total value) can afford to contribute more.

Platforms may also adjust pricing dynamically based on market conditions. Surge pricing in ride-sharing (charging riders more in peak times) is a form of dynamic pricing to balance demand and supply – it’s not exactly the platform fee but affects the split (often the platform still takes its percentage, but higher fares yield higher absolute commission). Some platforms temporarily reduce or waive fees to spur growth in a new market or category, then reinstate them. Promotional pricing (like Atlassian’s 95% revenue share to app developers in their first year) is used to attract innovation. As a platform owner, building pricing flexibility is important – you might start free, then introduce fees later (as Facebook did with ads, or open-source software projects do with paid cloud services), or start with one model and pivot (BlaBlaCar’s journey: tried ads and freemium, then found transaction fees with online payment worked best to reduce cancellations). The critical point is to monitor ecosystem health metrics (growth, activity, retention on each side) and be ready to adjust pricing to maintain a healthy equilibrium.

5. Alignment with Value Creation:
Lastly, good platform pricing strategy ensures that price correlates with value received by the participant. If users feel the platform’s cut or fee is too high relative to the value it provides, they will resist or churn. That’s why understanding price elasticity and doing value-based pricing analysis is key. For example, enterprise software marketplaces (like AWS Marketplace) can charge a relatively low commission (AWS takes around 3% for SaaS listings) because the value to software sellers of an AWS integration is high but AWS’s involvement is minimal beyond facilitation. Meanwhile, app stores justified 30% for years as covering distribution, payment, and global reach costs – though pressure is mounting to justify that rate as competitive and regulatory forces question it. Pricing also sends a brand message: platforms often want to be seen as partner-friendly. If a platform positions itself as developer-friendly, lowering fees (as Epic did against Apple, or Shopify did for app devs) is a strategic move to attract more partners. Conversely, some high-value platforms can command premium pricing (Apple can still enforce a 30% cut for large developers due to its massive user base). In all cases, communicate clearly what the fee pays for – e.g. Apple highlights the ecosystem, tools, and security it provides for the commission. Ensuring that participants feel they get value for the fees (or that the other side is paying their fair share) is vital for long-term sustainability.

Real-World Examples of Platform Pricing Models

Let’s look at specific companies and how they charge in their platform roles, to illustrate the diversity of approaches:

• Apple App Store (iOS/macOS) – Apple runs one of the most influential consumer software platforms connecting developers and users. Its pricing model is a 30% commission on app sales and in-app purchases by default. However, in 2021 Apple launched the Small Business Program reducing the commission to 15% for developers earning under $1 million per year. Once a developer exceeds $1M, the standard 30% rate applies for the rest of that year. (They can requalify for 15% if revenue falls back below $1M). Apple also charges an annual $99 fee for the developer program (a modest flat fee) and takes 15% (instead of 30%) on subscription renewals after one year, incentivizing developers to offer longer subscriptions. The rationale Apple gives is that its commission is the industry-standard for app distribution, funding the App Store’s operations, review process, and developer resources. In effect, Apple largely monetizes the developer side via revenue share, while users pay for apps but do not pay any additional “access” fee to Apple beyond the device itself. (Apple’s ecosystem profits also come from device sales, of course, which is another part of their platform strategy – hardware as an entry fee to the software platform.)
• Google Play Store (Android) – Google’s model converges with Apple’s in many ways: Google also traditionally took a 30% cut of app transactions. In 2021, Google announced it would take 15% on the first $1M of developer earnings per year (mirroring Apple’s small developer tier). Additionally, Google dropped its service fee for all app subscriptions to 15% from day one (previously it was 30% then 15% after a year).com. This was to encourage more subscription-based apps. Like Apple, Google is thus charging primarily the developer/content side, aiming to keep the Android user base large and engaged (users pay for paid apps or in-app content, but there’s no “fee to use Google Play” for users). The Play Store also charges a $25 one-time registration fee for developers (much lower barrier than Apple’s annual fee). Both Apple and Google have faced pressure from large companies and regulators to allow third-party payment systems or app stores, which is influencing their pricing – but as of 2025, their 15%/30% tiered commission structures remain core.
• Shopify – Shopify is a platform in two senses: it’s an e-commerce platform for merchants (online store owners) and also hosts an App Store for third-party developers to offer plugins/extensions to those merchants. Shopify monetizes merchants through a subscription (tiers like Basic $39/month, etc.) plus transaction fees if they don’t use Shopify’s own payment system. Merchants who use Shopify Payments (powered by Stripe) pay standard card processing fees (~2.9% + $0.30) but no extra surcharge; if a merchant uses an external gateway, Shopify adds an additional 0.5%–2% fee on transactions depending on plan – effectively encouraging use of its payment platform. Thus, from merchants, Shopify earns predictable subscription revenue and some payment revenue share. On the developer side, Shopify historically took a 20% revenue share of app sales on its App Store. In 2021, to attract more apps, Shopify announced developers keep 100% of their first $1M USD in annual app revenue, and 85% thereafter (i.e. 15% cut). This generous exemption (which reset each year) was a COVID-era relief and signaled Shopify’s attempt to be the “most developer-friendly” commerce platform. In 2025, Shopify revised this policy – instead of an annual reset, the 0% applies only to the first $1M lifetime earnings per developer, after which the 15% rate kicks in. Shopify reasoned that the yearly reset was mostly benefiting larger developers who hit $1M every year, rather than truly small newbies. Even after rolling back the reset, keeping only 15% (versus the old 20%) is relatively low and intended to keep the Shopify ecosystem vibrant. In summary, Shopify charges both sides of its platform in different ways: merchants pay subscription (and indirectly pay for payment processing), developers pay a revenue share – but Shopify calibrates these fees to not deter participation, even tweaking policies in response to developer feedback.
• Salesforce AppExchange – Salesforce’s platform for third-party enterprise apps (AppExchange) is a prime example of B2B ecosystem pricing. Salesforce offers partners two main models: ISVforce (apps that require a customer to have Salesforce) and OEM (apps that can be sold standalone on Salesforce’s platform). Salesforce takes 15% of revenue for ISVforce apps and 25% for OEM apps by default. For very high-volume apps (>$20M annual sales), the rate can drop to 10% (ISV) or 15% (OEM). Additionally, Salesforce has some fixed fees: a security review for apps ($2,700 initially, $300 annually) and a small annual listing fee ($150). These cover the cost of maintaining quality and trust on the platform. Notably, Salesforce’s model shows tiered revenue share (lower % for bigger players) to keep top partners happy. Salesforce does not charge end customers extra to use AppExchange per se – customers pay for the apps they buy, and the revenue share comes out of the partners’ sales. In essence, Salesforce monetizes the developer/partner side, aligning their success with Salesforce’s own. This is common in enterprise platforms, where the primary revenue is still from the core product’s licenses, and the ecosystem is monetized as an adjacent revenue stream (while simultaneously driving stickiness for the core CRM product).
• Atlassian Marketplace – Atlassian’s ecosystem (for Jira, Confluence, etc.) similarly runs on a revenue share model. Currently, Atlassian gives 85% of app revenue to cloud app developers (keeps 15%) and gives 75% to on-premise app developers (25% cut for Atlassian). Originally, the split was 70/30 for server apps, but Atlassian improved it to encourage cloud transition (now 75/25). They also had a special incentive where new cloud apps built on their Forge platform got 95% of revenue in the first year (i.e. Atlassian only took 5% initially) – a promotional strategy to spur innovation and adoption of their new app framework. These choices reflect a strategic use of pricing to drive platform goals (cloud adoption, more app development). Like Salesforce, Atlassian’s customers (the end-users of Jira, etc.) don’t pay Atlassian extra for using an app beyond the app’s price; Atlassian’s cut is taken from the partner’s revenue. This underscores that for software platforms, a revenue-share commission is a prevalent model – the platform provides the marketplace, billing, and customer base, and earns a portion of each sale for that facilitation.
• Amazon Marketplace (E-commerce) – Amazon’s third-party marketplace connects merchants with consumers. Amazon uses a combination of fees: referral fees (a commission on each sale, typically 15% in many categories), and a monthly seller subscription ($39.99 for Professional sellers). The referral fee percentage varies: some categories are lower (e.g. electronics 8% over a certain price), some slightly higher, but 8–15% is the usual range. There’s also a $0.99 per item fee for Individual (non-subscription) sellers. In essence, Amazon charges the sellers these fees; buyers pay nothing extra (except shipping or Amazon Prime for free shipping, which is a different value prop). Amazon additionally offers Fulfillment by Amazon (FBA) where sellers pay fulfillment fees, and Amazon makes money on warehousing and logistics services. Amazon’s platform strategy is to make selling as frictionless as possible (hence no listing fees for standard sellers, just pay when you sell) and to monetize each successful transaction plus value-added services. The massive scale and the value to sellers of accessing Amazon’s customer base justify the fees. For consumers, Amazon’s platform is free – they just pay product prices (which implicitly include those seller fees). It’s worth noting Amazon has built a huge advertising business on the marketplace too, where sellers bid on sponsored product slots – another layer of monetization (advertisers pay, users still free). Thus Amazon exemplifies multi-sided monetization: seller side primarily via commissions and service fees, buyer side indirectly via optional Prime memberships (for fast shipping and media) but not per transaction, and a tertiary advertiser side (brands and sellers paying for visibility).
• eBay – As one of the earliest digital marketplaces, eBay’s model has evolved but remains centered on seller fees as well. eBay traditionally charged a listing fee (insertion fee) plus a final value fee (commission). Today, insertion fees are often waived for a certain number of listings, and the final value fee is the main charge – roughly 12.9% of the sale price on most categories (slightly more or less in some cases) plus ~$0.30 per order. Buyers on eBay pay no fees to the platform. eBay also charges sellers for optional features (reserve price, bold listing, etc.) and offers promotions for fees if using eBay’s payment processing. The similarity to Amazon is clear (though Amazon’s system is more centralized whereas eBay’s is more peer-to-peer originally). Both demonstrate the pattern in consumer product marketplaces: charge the seller side a percent and perhaps a membership; keep it free or nearly free for buyers to maximize demand.
• Etsy – Etsy, a niche marketplace for handmade goods, charges sellers a $0.20 listing fee per item, and a 6.5% transaction fee on sales, plus payment processing fees ~3%. Again, buyers are not charged platform fees. This shows how even smaller specialized platforms combine a small listing fee (to ensure sellers don’t spam thousands of items with no cost) and a commission that is a bit lower than Amazon/eBay’s general 12-15%. Etsy also offers sellers optional advertising on the platform for additional cost. The theme remains: the producer side pays for access to the consumer side.
• Uber (Ridesharing) – Uber is a two-sided platform linking drivers (providers) and riders (consumers). Uber’s revenue comes from taking a “Service Fee” from drivers’ fares, plus in some markets a separate booking fee from riders. The driver-side commission is around 25% of each fare as a baseline. (This can vary – some Uber services have 20%, some newer drivers in certain cities have 25%, etc., and other factors like promotions can change the effective take rate.) Uber’s website explains this fee funds the app development, support, etc.. Riders see the full fare upfront; Uber then pays the driver their portion. In effect, riders pay the gross amount, drivers receive ~75%, Uber keeps ~25%. Additionally, Uber often charges riders a small fixed “booking fee” (a few dollars) on each ride which goes entirely to Uber to cover safety/operations – this is a way to earn a bit more from the rider side without it being a percentage. The combination means both sides contribute: drivers give up a fraction of earnings, riders pay slightly more than just the mileage/time (the extra goes to Uber). Uber and similar platforms (Lyft, etc.) leverage this model at scale – millions of microtransactions with a percentage fee. The challenge they’ve faced is driver pushback if the effective commission feels too high (especially when considering drivers also bear vehicle costs). Indeed, some calculations show that at times the platform’s cut can exceed 25% (if surge pricing or minimum fare dynamics apply). Uber must balance taking enough to be profitable with leaving enough for drivers to remain on the platform. This balance of take rate vs. participant earnings is critical in any platform – take too much and you undermine the ecosystem’s ability to sustain itself.
• Airbnb (Home Sharing) – Airbnb historically used the split fee model: guests pay roughly 5–15% on top of the reservation (Airbnb calls this a guest service fee, it varied with booking amount, averaging around 14%), and hosts pay ~3% from their payout.com. For example, a guest might pay $114 on a $100/night booking, the host gets $97, Airbnb keeps $17 total (which is 14% of the gross). This split-fee was simple for individuals. In recent years, Airbnb introduced a host-only fee option (mandatory for hotels and some pro hosts): the guest pays no service fee, and the host pays Airbnb 15% of the booking (or up to 16% with some stricter cancellation policies). This “transparent pricing” was to cater to property managers who preferred to bake the fee into the price shown to guests. Either way, Airbnb’s take rate is around 15-20% of the transaction in total. The platform also earns from selling add-ons like experiences (with similar commission) and now an optional insurance fee. Airbnb demonstrates a case of initially charging both sides a little, then evolving to offer different pricing structures for different segments. Compared to Uber, Airbnb’s cut is a bit lower percentage-wise and is partially borne by the demand side. This likely helped Airbnb scale supply early (hosts saw only a tiny fee), while guests might not have been as sensitive to a service fee on top since the total cost often was still lower than hotels. It’s a good example of multi-sided pricing that shares costs and the ability to shift who pays as the business matures (Airbnb could later ask hosts to pay more when it had sufficient demand leverage).
• Stripe (Payments Platform) – Stripe provides payment infrastructure as a platform primarily for developers and online businesses. Its pricing is a classic usage-based model: Stripe charges 2.9% + $0.30 per successful card transaction (in the US; rates vary slightly internationally). There are no setup or monthly fees for the basic service. Essentially, the merchant (Stripe’s user) pays this fee per transaction; the end-customer sees just the price of the product (the fee is embedded in merchant’s costs). Stripe thus monetizes by aligning with the flow of money – if a business grows and processes more volume, Stripe earns more proportionally. Stripe also offers volume discounts and custom pricing for large clients, and charges additional fees for certain products (e.g. +0.5% for Stripe Billing invoicing, small fees for Instant payouts, etc.). But the core idea is a platform for payments where only one side (the business) pays, and it pays as a function of value received (processed payments). This model has made Stripe very popular because it’s pay-as-you-go and scales with the user’s success. It lowers barrier to entry (no fixed fees) which is crucial for startups – again reflecting how removing friction on one side (business adoption) by only charging per usage helped Stripe become a default choice for developers. Competing payment platforms (Braintree, PayPal) have similar fee structures, indicating a standard in that domain.
• Visa/Mastercard (Card Networks) – The credit card networks are a less visible platform, but indeed they connect cardholders (via issuing banks) and merchants (via acquiring banks). Their pricing is complex (interchange fees, network assessments) but effectively, merchants pay around 2%–3% per transaction in fees which are split among the issuing bank, the network, and the acquiring provider. Cardholders often pay no fee and in fact receive rewards funded by those fees. This is a textbook two-sided market strategy: charge the side that benefits most financially (merchants gain sales by accepting cards) and subsidize the consumer side to maximize card usage. It’s a reminder that platform pricing predates digital – malls charged rent to stores, not shoppers; Yellow Pages charged businesses for ads, not consumers for searching – the pattern of asymmetric pricing to spur network creation is common. In the card example, the “price” to consumers is actually negative when rewards/cashback are given, showing how far subsidy can go when the other side can bear it.
• Facebook / Meta Platforms – On social media platforms (Facebook, Instagram, WhatsApp, etc.), the user side is entirely free. The platform monetizes via advertisers who pay for targeted ads. This has been extraordinarily profitable because of scale – billions of users at price $0 creates an attractive audience. Advertisers’ willingness to pay grows with the user base and the data available for targeting. The pricing model for ads is typically an auction-based system (cost per click or impression), not a fixed “fee”, but essentially the advertisers are the paying customers. This model can be thought of as a multi-sided platform connecting consumers, content creators, and advertisers. The content creators (e.g. media companies, influencers) also typically don’t pay to use the platform, but some platforms share revenue with them (YouTube shares a portion of ad revenue with video creators, usually 55% to the creator, 45% to YouTube). In that sense, YouTube charges advertisers, then splits that revenue with the “supply side” of content – a hybrid of ad model and revenue share model to incentivize content creation. The overarching point is that monetization can come from a third party if direct charging would impede growth or engagement. This works best when you have a product people use frequently and for long periods (so ads can be served) and when the presence of ads doesn’t drive away users (a balance of user experience).
• AWS Marketplace and Cloud Platforms – Cloud providers like Amazon AWS, Microsoft Azure, and Google Cloud have marketplaces where third-party software (SaaS or virtual machine images) can be sold to their cloud customers. Interestingly, AWS Marketplace’s fees are relatively low: for SaaS listings sold through AWS, Amazon takes only 3% of the contract value as a listing fee. For software that’s packaged as machine images (AMI) or containers, the fee is higher (20%), since AWS is handling more infrastructure billing in that case. They also have a sliding scale for large enterprise deals (private offers) – fees drop to 2% or 1.5% for big contracts. This low take rate (compared to, say, app stores at 30%) reflects that in B2B enterprise sales, the margins are different and AWS likely makes money from the underlying cloud usage anyway. The marketplace fee is more about covering costs and encouraging vendors to list. It also shows how competitive pressure can influence platform fees: these cloud marketplaces compete with direct sales and other channels, so keeping fees low encourages adoption. Azure and Google Cloud similarly charge around 3–5% for marketplace transactions. Essentially, the platform in this case (cloud provider) is subsidizing the ecosystem to make their cloud more attractive – the real money for them is in compute, storage, etc. This is a case where the platform’s primary revenue is elsewhere, and the marketplace is an add-on; thus pricing it high could deter participation and hurt the core business. It underscores how a platform must consider the total value of an ecosystem participant (not just revenue from fees but also indirect revenue).
• Upwork and Fiverr (Freelancing Platforms) – These labor marketplaces have an interesting two-sided structure. Upwork charges freelancers a percentage of their earnings (20% of the first $500 with a client, 10% of the next ~$9,500, and 5% beyond that – encouraging long-term relationships) and it charges clients a processing fee (~3% on payments) or a flat monthly fee for a premium plan. Fiverr charges sellers (freelancers) 20% on all earnings and charges buyers a service fee (~5% of purchase, with a minimum $2). In both, the provider side pays the larger cut, but the buyer side isn’t entirely free either – which helps the platforms earn more per transaction. They justify the freelancer fee as covering marketing, payment protection, etc. The small buyer fee is positioned as a processing/service fee. This is a model of charging both sides, but not equally: roughly speaking, 80% of revenue comes from the supply side fee, 20% from the demand side fee in these cases. It can work when the value delivered to the paying sides is clear – freelancers get access to clients and guaranteed payment, clients get access to talent and platform conveniences. If either side had great alternatives or less need, the platform might have to drop fees on that side to keep them.

Each of these examples highlights how platform pricing is tailored to the specific ecosystem: consumer vs. enterprise, digital vs. physical, high-frequency vs. occasional use, etc. Nonetheless, patterns emerge, most platforms monetize where the money flows or where the captive value is, and strive to keep at least one side of their network as friction-free as possible.

TL; DR

Pricing for platforms is a balancing act that must consider economics, user experience, and strategic positioning. Unlike one-sided businesses, platforms succeed by creating value for multiple parties simultaneously, and their pricing structure must reflect that interdependence. Key takeaways from this exhaustive survey include:

• There is no one-size-fits-all model – successful platforms often innovate new pricing schemes or combine models to fit their market. The pricing strategy should evolve with the platform’s growth stage and competitive context. Early on, maximizing network size (even at the expense of revenue) is often wise; later, capturing value sustainably becomes the focus.
• Charge the side(s) that can bear it (those who get the most value or have fewer alternatives) but beware of overcharging. Excessive fees can repel the very users that make the network valuable or drive them to find workarounds off-platform. The art of platform pricing is often in identifying who really needs the platform more, and setting a price that reflects that value-add without feeling extractive.
• Use pricing to shape the network: lower it to remove friction, tier it to incentivize desirable behavior (e.g. volume discounts, loyalty), and even consider subsidies or negative pricing to kickstart growth or correct imbalances. Pricing is a strategic lever to balance supply and demand (think of surge pricing, or offering discounts to attract the lagging side of a marketplace). It’s not just about revenue – it’s a tool to engineer network effects.
• Monitor and adjust: Data is crucial. Platforms should continuously analyze how participants respond to pricing changes – do conversion rates drop if a fee is introduced? Does supply grow if we cut their commission? As Launchworks experts noted, leaders need “pricing confidence” which comes from understanding both the direct impact on payers and the indirect effects on the whole ecosystem. This often means running experiments or phased changes and gathering feedback.
• Competitive and Regulatory landscape: Platform pricing does not happen in a vacuum. If competitors offer a more favorable fee structure to one side, a platform may need to match it (e.g. Epic Games store undercutting Steam’s 30% fee with 12%, or Shopify reducing app fees to attract developers away from other ecosystems). Regulators are also scrutinizing platform fees – for instance, app store commissions and payment exclusivity have come under legal challenges. Platform owners should be proactive in justifying their fees by emphasizing the value provided, and perhaps preemptively adjusting overly skewed pricing that could invite backlash. In some cases, more open or flexible pricing (or allowing third-party alternatives) can be a competitive advantage or at least reduce regulatory risk.

Pricing when “you’re the ecosystem” is about finding the sweet spot that fuels growth and engagement on your platform, while capturing enough value to be profitable and invest in the ecosystem. It often requires an exhaustive understanding of user segments and their needs. By examining what successful platforms across industries have done – from subsidizing one side entirely, to charging both sides in clever ways, to inventing new fee structures – platform strategists can gather a toolkit of approaches. Ultimately, the best pricing strategy aligns with the platform’s core value creation: it should reinforce the network effects (not work against them), reward the most beneficial participants, and ensure the platform’s long-term viability as an orchestrator of value. With thoughtful, flexible pricing design, a platform can grow into a thriving ecosystem and truly charge in proportion to the unique value it delivers as the ecosystem hub.

Sources: The insights and data above were compiled from a range of up-to-date sources, including industry whitepapers, company documentation, and expert analyses on platform monetization. Notable references include MIT Sloan research on multisided platforms, interviews with platform strategists at Launchworks, and official figures from platform companies (Apple, Shopify, Salesforce, Amazon, Airbnb, Uber, Stripe, and others). These examples illustrate the principles in practice, providing a foundation for developing pricing strategies when “you’re the ecosystem.”