PDF Summary:Platform Revolution, by Geoffrey G. Parker, Marshall W. Van Alstyne, and Sangeet Paul Choudary

Have you ever wondered how platform businesses like Spotify and Amazon became so successful? In Platform Revolution, a team of economists and businesspeople—Geoffrey G. Parker, Marshall W. Van Alstyne, and Sangeet Paul Choudary—offer some insight: Platform businesses harness their unique characteristics to drive competitiveness and transformation, resulting in profound economic, societal, and political impacts. Parker, Van Alstyne, and Choudary also highlight the distinct skill set required to operate such a business and provide practical tips for aspiring platform entrepreneurs.

In our guide, we’ll compare the platform business model to the traditional business model known as a pipeline, explain why platforms are becoming dominant, and explore their wide-ranging effects. We’ll also outline essential design and management strategies for platform developers. Our commentary will provide additional insights on economic concepts, discuss tips from business experts, and explore real-world applications of Platform Revolution principles.

(continued)...

Parker, Van Alstyne, and Choudary stress the importance of highlighting your platform’s central purpose and sidelining additional features when you design your platform’s software. They explain that as your platform grows, users’ needs may evolve, and your platform should develop additional features to meet those needs. However, you should use software tools to make your central purpose visually dominant and easy to find (for example, by putting it on users’ homepages) while making additional features accessible through a navigation menu. This ensures that your central purpose doesn’t get lost or overshadowed and clarifies the value your platform offers to users.

To illustrate, consider the design of X. When you open the app, you see tweets from people you follow and recommended tweets—this makes sense since the platform’s central purpose is to connect users via tweets. If you want to take advantage of other features, like X spaces, you can navigate to them using the menu. This design allows users to access additional features without adding distracting clutter to the main feed.

(Shortform note: In Dotcom Secrets, Russell Brunson characterizes this strategy as a sales funnel, or a standardized website layout that puts the product you’re selling front and center and easily guides customers through each step of making a purchase. Sales funnels are key to building a successful e-commerce website because they streamline the buying process, which translates into an optimized user experience and increased sales. Although Brunson is primarily writing about pipeline e-commerce businesses whose central purpose is to sell products, the sales funnel concept can easily be adapted for all types of platforms with various core purposes: For example, X uses the concept to streamline its central purpose (connection via tweets).)

Choose Your Participants

As you design your platform, you must make choices about which participants you want to invite to your platform and what they’ll be allowed to do. Parker, Van Alstyne, and Choudary primarily focus on two types of participants: users and external developers.

Attracting Users to Your Platform

Parker, Van Alstyne, and Choudary stress that to attract users to your platform, you must ensure your platform has obvious inherent value. One way to accomplish this is by ensuring that users can use your platform to make valuable connections in a straightforward, pleasant way—if it’s a hassle, users will lose interest. The authors list three strategies to attract users: First, promote easy access to your platform (for example, don’t charge a fee or require an ID for users to join) to increase your user base. Second, balance easy access with selectivity—don’t let low-quality users enter or remain on your platform. Third, utilize user data, like reports of abuse and activity patterns, to inform your decisions about selectivity and recommend high-quality connections.

(Shortform note: The authors’ recommended strategies for attracting users can help your platform garner attention, retain a high-quality user base, and facilitate valuable connections, but experts suggest these effects are only worthwhile if your platform is also easy to use. A platform’s usability is determined by its technological design and includes aspects such as simple forms, clear and aesthetically pleasing text, and search functionality. Usability is one of the key characteristics consumers look for when they’re considering whether to adopt a new platform, so experts say developers should design platforms with key usability considerations in mind.)

New platforms also face a unique challenge—determining how to attract producers when there are no consumers yet, or vice versa, since the platform is only valuable if both types of users are active. The authors discuss several strategies you can use to address this challenge encouraging you to mix and match strategies to find the best approach for your platform:

Strategy #1: Piggyback on a pipeline. If you build a platform that has obvious value because it’s comparable to pipeline businesses in the same market, it’ll be easier to convince users to sign up. For example, Uber has obvious value as an alternative to traditional taxi companies.

(Shortform note: If you’re struggling to find a comparable pipeline to piggyback on, an alternative strategy is to partner with a pipeline and tap into its user base. For example, experts explain that while pipeline-based grocery delivery services have failed in the past, the grocery delivery platform Instacart has succeeded because it leveraged partnerships with existing grocery stores (pipelines) to create a network consisting of their shoppers.)

Strategy #2: Generate buzz with a small initial network. If you capture a very small percentage of potential users and gain publicity by proving your value, it’ll be easier to scale up later. For example, Pinterest used an invite-only model to create exclusivity and generate early interest before widening access.

(Shortform note: Uber executive Andrew Chen recommends Strategy #2 in The Cold Start Problem. Chen explains that you can attract your initial network by targeting an extremely niche market and ensuring your platform meets users’ unique needs. Chen also emphasizes that you need both producers and consumers for this approach to work, so you should cater to whichever type of user is more difficult to attract until you reach a point of stability. Then, you can repeat this process to attract other niche networks, gradually building your platform’s overall network. Eventually, you’ll accumulate enough users to reach exponential network growth.)

Strategy #3: Recruit key consumers or producers, but not both. If you retain a subset of clearly valuable users, users of the other type will soon flock to your platform. For example, the celebrity video message platform Cameo started off with minor-league celebrities to draw in the fans who wanted to hear from them.

(Shortform note: How do you convince producers to join your platform when there aren’t any consumers, or vice versa? One strategy is to pay them to use the service—Cameo’s CEO says he paid its first star $150 an hour, and as a startup, Uber paid drivers to stay online so that riders could have access to a reliable service. This requires access to a sum of cash upfront—but you can make up your losses later, when money is flowing between consumers and producers on your platform and users are more willing to pay to use it.)

Strategy #4: Choreograph an exemplary product launch. If you accumulate exemplary offerings from key producers, consumers will be more inclined to engage with your platform from the beginning, fostering trust and loyalty. For example, Spotify started out with a limited library of streamable music to attract consumers and then expanded its library as it grew.

(Shortform note: In Launch, entrepreneur Jeff Walker offers three tips you can use to optimize your product launch: First, stretch your launch out over days or weeks. This gives users a chance to discuss your platform with others, generating good publicity. Second, use a countdown with a definite endpoint to engender a sense of urgency and excitement among your users. This can help motivate users to join your platform. Finally, generously offer no-cost value to your users before you ask them to pay for something (for example, by launching a free version of the platform first). This can build users’ trust in your platform, making them more willing to spend money down the line.)

Orchestrating External Developer Relationships

In addition to attracting users, you must carefully orchestrate your platform’s relationships with external developers, the software engineers and data aggregators that develop extensions, applications, and advertising strategies to help users get the most value from your platform. Parker, Van Alstyne, and Choudary stress that third-party developers can siphon money from your platform if their creations are productive, so it’s important to acquire companies or creations that add significant value to your platform. For example, that’s why Twitter acquired TweetDeck, a third-party application that helped users customize their experience. It’s also important to manage data acquisition and sharing ethically to maintain user trust.

You should have a plan for orchestrating these relationships in mind as you design your platform, but the authors recommend keeping an open mind: As the market transforms over time, it may behoove you to allow external parties more or fewer permissions to work with your platform.

Case Study: Reddit’s Management of External Developer Relationships

In 2023, the platform Reddit ignited controversy by updating the terms of its external developer relationships (we’ll call these EDRs). These updates made it significantly more expensive for external developers to work with Reddit, which led several to shut down their apps. Many Reddit users protested or left the platform because of this policy, but Reddit persisted in an attempt to prevent artificial intelligence (AI) companies from training large language models (LLMs) like ChatGPT on Reddit’s data without compensation. In 2024, Reddit made its first deal with an AI company, Google, which makes user data available to the company for training purposes. The deal’s announcement also resulted in user backlash.

Let’s discuss how Reddit’s EDR management strategy aligns with—and departs from—the three strategies Parker, Van Alstyne, and Choudary highlight.

The authors recommend that platforms acquire valuable external developers. In updating the terms of its EDR policies, Reddit took the opposite approach: severely limiting the ability of external developers to add value to the platform. This backfired in some ways—for example, some users left the platform because the affected developers’ extensions made Reddit more accessible, and losing them rendered the platform unusable. The decision could also have negative financial ramifications, given its damage to network effects, but these should be ameliorated by the profits Reddit makes via its deal with Google.

The authors recommend that platforms ethically manage data acquisition and sharing; ambiguity about the meaning of “ethical” complicates this suggestion. At a minimum, Reddit’s privacy policy appears to provide legal cover for its deal with Google, since it states that Reddit can share users’ data in certain contexts. Furthermore, academics already utilize social media users’ data to conduct research since it’s public, and data aggregators can legally mine public online data without seeking users’ affirmative consent, which suggests the development of a new minimal ethical standard. However, this standard is often contested by those who feel their privacy—and in some cases, their property rights—are being infringed.

Finally, the authors recommend that platform owners keep an open mind to cope with market transformations. Reddit clearly followed this piece of advice—the advent of generative AI, a type of system that can create new content, is quickly revolutionizing the global market in unexpected ways. Since generative AI is trained to create content based on its statistical analysis of massive amounts of data, data mines are becoming more and more valuable—so it makes good financial sense for Reddit to capture a percentage of the value it provides to AI trawlers.

Decide How to Turn a Profit

The authors explain that profiting from your platform can be difficult because most of the obvious profit-making strategies, like charging users to join, can lower your platform’s value by impeding easy access and decreasing the size of your network. For this reason, many platform designers elect a free or “freemium” (free with paid premium upgrades) model until the network is sufficiently built up. Then, they can consider implementing profit-making strategies. If you have a similar plan for your platform, note that charging for benefits that were once free—like access to basic features—is a surefire way to make users feel exploited, which could lead to an exodus of users. Instead, create additional value (like exclusive content or an ad-free experience) and charge for that.

(Shortform note: Although the authors warn that charging users for benefits that were once free can spur users to leave your platform, recent data suggests the opposite. In 2023, Netflix introduced measures to prevent password-sharing, which sparked user ire but paid off for the company in the form of nearly 10 million new subscriptions. In 2024, Amazon introduced ads to its streaming service, offering users the option to pay to get rid of them. This move was also unpopular with users, but experts suggest that it will be profitable and other streaming services are likely to follow suit. Users may tolerate such changes if they view a platform as indispensable, as streaming services have become in the entertainment sector.)

Effective profit-making strategies capitalize on user benefits—like producers’ and consumers’ access to each other—without motivating users to leave the platform or use it less frequently. This requires understanding the spending ability of each user market—for example, ultra-profitable producers may have more money to work with than consumers with little disposable income.

(Shortform note: There are several ways to ensure your platform is affordable for users, given their spending ability. For example, you might conduct market research by collecting and analyzing data that capture your targeted users’ spending habits, or you might price-match with competing platforms.)

Effective profit-making strategies include:

• Upgrading selectivity services for a fee. For example, you might offer tiered subscriptions that give users access to higher-quality offerings. (Shortform note: The most common form of this strategy is to offer three tiers: good, better, and best.)
• Billing users for access to each other. For example, you might let producers pay to find out more about consumers’ desires so they can tailor their offerings. (Shortform note: To address users’ data privacy concerns, you might ensure this is opt-in only.)
• Taking a percentage of payouts. For example, you might impose a service fee on consumers or take a share of producers’ profits for each transaction. (Shortform note: To avoid user backlash, experts suggest imposing fees that are sensible and predictable.)

Platform Management Principles

Once you launch your platform, you need to manage it carefully to ensure its success. To help you accomplish this, we’ll cover three key management principles in this section: strategic platform administration, tailored performance measurements, and clever competitive tactics.

Strategic Platform Administration

A strategic platform administration system can help you maintain control of your platform’s quality as it grows. This is important for two reasons: First, platform experiences are an increasingly significant part of life for countless people, and they have real-world implications. For example, banning a user from a social media platform could impact their ability to contact their loved ones. To keep users happy, Parker, Van Alstyne, and Choudary state that your administration must be straightforward, clear, judicious, and fair.

(Shortform note: Controversy about an administrative practice called “shadowbanning,” where social media platforms restrict the visibility of a user’s content without informing them, illustrates the importance of a strong platform administration system. Research suggests that users find shadowbanning frustrating and unfair, since it’s applied inconsistently and often with a seemingly political bent, which has fueled a crisis of trust in social media platforms. Because of this distrust, social media users have collectively invented a new lexicon called “algospeak,” which replaces words that might get you shadowbanned with made-up alternatives (like “seggs” for “sex”). Algospeak is an attempt to skate around the rules, undermining platforms’ authority.)

Second, Parker, Van Alstyne, and Choudary argue that strategic administration can increase the value of your platform. One reason for this is that poor administration can lead to “market failures,” an economic phenomenon that occurs when producer-consumer connections are unsuccessful or exploitative or result in harm to others. A preponderance of market failures decreases the value of your platform by diminishing its reputation, reducing user engagement, and decreasing revenues.

(Shortform note: As Niall Ferguson argues in The Square and the Tower, US platforms are largely unregulated, which means that users are liable to encounter a variety of market failures. At worst, these include dangerous producers and products, including social media trends that encourage harmful behaviors, platform-mediated exposure to violence, and purchasable items that cause fire and injury. However, many platforms that produce market failures remain valuable (for example, Craigslist was worth at least $3 billion in 2017, despite its link to over 100 murders). As we discussed with regard to seemingly exploitative monetization strategies earlier, users may be willing to take on huge risks for platforms if they deem them indispensable.)

The authors explain that strategic administration systems prevent market failures in four ways: First, they lay down hard-and-fast rules with clear consequences (for example, you might prevent market failures by banning users who incite violence). Second, they proactively promote desired behaviors (for example, promoting engagement by notifying users of trending activity and prompting them to contribute). Third, they develop software that deters market failures (for example, an automatic moderation feature that removes posts and listings made by bots). Fourth, they use economic tools to shape supply and demand and minimize risk (for example, they might eat the costs of market failures like fraudulent sales, fostering user trust and participation).

(Shortform note: As you decide which market failure prevention strategies to use, experts say you should ensure that the strategies cover four key areas of concern: First, your strategies must cover general concerns like privacy and security to maintain a basic level of trust and safety on the platform. Second, to protect your business interests, they must encourage profitable behaviors and prevent issues that could cost your platform money. Third, they must shield your company from legal harm by promoting responsible technology use. Fourth, they must address platform-specific concerns (for example, it may be more vital to address trolling and harassment on a platform like LinkedIn, whose central purpose requires professionalism.)

Tailored Performance Measurements

Parker, Van Alstyne, and Choudary say that many traditional performance measurement tools can’t account for network effects, so they won’t be useful for evaluating the financial health of your platform. They recommend using metrics that capture one of three key indicators of platform performance, depending on which stage of life your platform is in.

(Shortform note: Although traditional performance measurement tools may not work for your platform business, some traditional performance measurement strategies may still apply. For example, in Traction, business consultant Gino Wickman recommends that you compile relevant financial information in a weekly scorecard instead of waiting for monthly or quarterly profit and loss statements. This strategy could keep your finger on the pulse of your platform’s performance, enabling you to respond quickly to surfacing challenges and more proactively strategize for the future.)

For new platforms, measure the value you offer to users. Quantifying this value helps you prove that your platform will grow and succeed. There are many ways to measure value to users. For example, Parker, Van Alstyne, and Choudary recommend evaluating user satisfaction by tracking users’ activity and measuring their trust that the benefits of participating in your platform outweigh the risks.

(Shortform note: Measuring value to users informs managers about your platform’s performance, and it can also be used in communications with users to help them see the value you provide. For example, this is the strategy behind Spotify Wrapped, a personalized summary of users’ yearly activity. Spotify Wrapped reminds users of the value they derive from the platform, and since it’s designed to be shared across social media, it can inform prospective users of the platform’s value, too, which may prompt them to join. Since it’s a great marketing tool, sharing value measurements has become widespread among a variety of platforms.)

For scaling platforms, determine the equilibrium between producers and consumers and the value each user type offers. Parker, Van Alstyne, and Choudary explain that if there’s a mismatch in the number or quality of producers and consumers, you must implement strategies to ensure users are still getting value from the platform. To illustrate, imagine there’s been an influx of producer bots on your platform, resulting in unfulfillable sales listings. You’d need to remove the bots to keep consumers happy.

(Shortform note: Economists have long recognized the need for equilibrium between producers and consumers, but historically, in pipeline contexts, equilibrium has been measured by evaluating the price of a product. In this context, we’re more interested in the number and quality of each user type because the value of the platform—the product in question—is determined by the engagement between producers and consumers, not the price of their services or the price of the platform.)

For well-established platforms, study users’ activity to determine opportunities to improve your value to users. For example, this is why Facebook began Facebook Marketplace—it noticed its users were using the platform to buy, sell, and trade. The authors explain that determining opportunities to improve value helps you remain competitive—a necessity we’ll talk about in the next section.

(Shortform note: In Start With Why, Simon Sinek illustrates the importance of focusing on existing users. There are two kinds of consumers: early adopters, who latch onto new products immediately, and latecomers, who require more convincing. Many innovators believe that they should prioritize reeling in latecomers because they tend to outnumber early adopters, but Sinek says this is a trap: You probably won’t persuade latecomers, but early adopters can serve as brand ambassadors who will reel in convertible latecomers for you. This explains why studying existing users’ activity to determine opportunities to offer them more value helps you remain competitive—it’s more effective and resource-efficient than converting new users to your platform.)

Clever Competitive Tactics

According to Parker, Van Alstyne, and Choudary, traditional competitive strategies, like those described by Michael Porter in Competitive Strategy, don’t fully account for the nature of competition among platforms. One reason for this is that platforms disrupt markets by leveraging network effects. For example, consider how Spotify fundamentally changed the music market by introducing a streaming library, which shifted the consumption model from ownership to access, reshaping supply and demand. Instead of trying to steal artists from their record labels (the old competitive mode in the music industry), Spotify collaborates with a large network of stakeholders, including artists and their labels, to negotiate deals that it claims create value for everyone.

(Shortform note: For decades, business experts recognized Porter’s Competitive Strategies as the quintessential framework for strategic management. Although experts (including Parker, Van Alstyne, and Choudary) agree that some of Porter’s wisdom is outdated because it doesn’t account for network effects, some of his advice may still be helpful for platform competition. For example, Porter explains that analyzing your competitors’ strategies can help you anticipate and counter their strategic moves, like in a game of chess. This tactic can give platform managers important insights—for example, managers could look at pricing, innovations, and other strategic decisions by rival platforms and think through how to counter or match them.)

The authors recommend a variety of competitive strategies you can use to enhance your platform’s value. We’ve already covered three strategies: acquiring external developers’ valuable creations, tracking user data to identify opportunities to improve value, and partnering with competitors to share value, as Spotify did. Let’s discuss three more:

Offer better quality than your competitors. This can help you attract new users who are dissatisfied with your competitors or to gain the trust of existing users. For example, you might set yourself apart from your competitors by offering better customer service, smoother in-app performance, or extra features.

(Shortform note: In Ten Types of Innovation, Larry Keeley, Ryan Pikkel, Brian Quinn, and Helen Walters explain that innovation can help you offer the best possible quality to users. According to these experts, there are ten types of innovation. Some may be particularly useful for platform managers: for example, these include product innovations that improve your platform’s basic functionality or ability to interface with other softwares and capability innovations, like those that harness new technologies to help users achieve what was previously unattainable.)

When a competitor offers a form of value you don’t, develop your own alternative. Innovation helps you stay relevant, which diminishes your chance of losing users to another platform. For example, Instagram developed Reels on the heels of TikTok’s success, which helped it recapture the interest of a percentage of TikTok users.

(Shortform note: In Competing Against Luck, business consultant Clayton Christensen advises strategists on how to identify and capitalize on opportunities to create consumer value. Christensen says that people buy products to accomplish specific tasks, so to meet their needs and sell them your product, you must have a deep understanding of the tasks your target market wants to accomplish. In a platform context, you can leverage Christensen’s guidance to identify the key tasks your competitors help users accomplish and innovate accordingly to meet similar needs.)

Discourage “multihoming,” or the use of multiple similar platforms, by enforcing exclusivity. This way, you can create a specific brand experience that appeals to your users more than competitors’ brand experiences, which incentivizes user loyalty. For example, Playstation forms exclusive deals with game developers to incentivize users to choose its platform over competitors.

(Shortform note: To reap the benefits of an enforced exclusivity approach, platforms must reliably deliver the specific brand experience they’ve promised their users—otherwise, users might become dissatisfied and fall away. The makers of Playstation encountered this problem following the release of the fifth generation of the platform, the PS5: The company failed to secure enough exclusive deals with game developers, and as a result it anticipates slower sales, which may hurt the company’s bottom line.)