Skip to content
Branch: master
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
1057 lines (779 sloc) 79.6 KB

Platform Revolution Chapter Summaries

How networked markets are transforming the economy and how to make them work for you.

Title Author Rating Type
Platform Revolution Geoffrey G. Parker ⚪️♂, Marshall W. Van Alstyne ⚪️♂, Sangeet Paul Choudary 🔵 ⭐️⭐️⭐️⭐️ Really helpful if you work on platforms 🤓

Symbol Meaning
💼 Case study
🔖 Definition
🔑 Key point

Notes from:

Chapters 1 & 2: @kytrinyx & @bkeepers

Chapters 3 & 4: @johndbritton & @wilhelmklopp

Chapters 5 & 6: @JessRudder & @mscoutermarsh

Chapters 7 & 8: @pifafu & @venetucci

1. TODAY - Welcome to the Platform Revolution

The book argues that the "platform" is "a new business model that uses technology to connect people, organizations, and resources in an interactive ecosystem in which amazing amounts of value can be created and exchanged."

A platform is a business based on enabling value-creating interactions between external producers and consumers. The platform provides an open, participative infrastructure for these interactions and sets governance conditions for them. The platform’s overarching purpose: to consummate matches among users and facilitate the exchange of goods, services, or social currency, thereby enabling value creation for all participants.

So…people can buy and sell stuff (for various definitions of "buy" and "sell" and "stuff") from each other verses from a single supplier, like in this old model:

The traditional system employed by most businesses is one we describe as a pipeline. By contrast with a platform, a pipeline is a business that employs a step-by-step arrangement for creating and transferring value, with producers at one end and consumers at the other. A firm first designs a product or service. Then the product is manufactured and offered for sale, or a system is put in place to deliver the service. Finally, a customer shows up and purchases the product or service.

In the world of platforms […] Rather than flowing in a straight line from producers to consumers, value may be created, changed, exchanged, and consumed in a variety of ways and places, all made possible by the connections that the platform facilitates.

"Platforms beat pipelines":

  • "…because platforms scale more efficiently by eliminating gatekeepers.” - traditional book publishers (manual selecting authors, long lead times) vs amazon self-publishing; "One reason is that the pipelines rely on inefficient gate keepers to manage the flow of value from the producer to the consumer”
  • “…because platforms unlock new sources of value creation and supply.” - traditional hotels (physical property) vs AirBnB and the sharing economy and (dynamic supply; “growth is no longer constrained by the ability to deploy capital and manage physical assets” and “Supply now unlocks spare capacity and harnesses contributions from the community which used to be only a source of demand.”
  • “…by using data-based tools to create community feedback loops.”

There's also this weird side-effect where “platforms invert the firm.”

because the bulk of a platform’s value is created by its community of users, the platform business must shift its focus from internal activities to external activities.”


  • A platform’s overarching purpose is to consummate matches among users and facilitate the exchange of goods, services, or social currency, thereby enabling value creation for all participants.
  • Because platform businesses create value using resources they don’t own or control, they can grow much faster than traditional businesses.
  • Platforms derive much of their value from the communities they serve.
  • Platforms invert companies, blurring business boundaries and transforming firms’ traditional inward focus into an outward focus.
  • The rise of the platform has already transformed many major industries—and more, equally important transformations are on the way.
More quotes from chapter 1

In the traditional publishing industry, editors select a few books and authors from among the thousands offered to them and hope the ones they choose will prove to be popular. It’s a time-consuming, labor-intensive process based mainly on instinct and guesswork. By contrast, Amazon’s Kindle platform allows anyone to publish a book, relying on real-time consumer feedback to determine which books will succeed and which will fail.

The elimination of gatekeepers also allows consumers greater freedom to select products that suit their needs.

“Uber, the world’s largest taxi company, owns no vehicles. Facebook, the world’s most popular media owner, creates no content. Alibaba, the most valuable retailer, has no inventory. And Airbnb, the world’s largest accommodation provider, owns no real estate.” The community provides these resources.

One implication is that growth can be much faster for Airbnb or any rival platform than for a traditional hotel company since growth is no longer constrained by the ability to deploy capital and manage physical assets.

Whereas the leanest traditional businesses ran on just-in-time inventory, new organizational platforms run on not-even-mine inventory.

Thus, platforms disrupt the traditional competitive landscape by exposing new supply to the market. Hotels that must cover fixed costs find themselves competing with firms that have no fixed costs. This works for the new firms because there is spare capacity that can be brought to market through the assistance of the platform intermediary. The sharing economy is built on the idea that many items, such as automobiles, boats, and even lawnmowers, sit idle most of the time.

Strategy has moved from controlling unique internal resources and erecting competitive barriers to orchestrating external resources and engaging vibrant communities. And innovation is no longer the province of in-house experts and research and development labs, but is produced through crowdsourcing and the contribution of ideas by independent participants in the platform.

But platform firms emphasize ecosystem governance more than product optimization, and persuasion of outside partners more than control of internal employees.

2. NETWORK EFFECTS: The Power of the Platform

Network effects refers to the impact that the number of users of a platform has on the value created for each user. Positive network effects refers to the ability of a large, well-managed platform community to produce significant value for each user of the platform. Negative network effects refers to the possibility that the growth in numbers of a poorly-managed platform community can reduce the value produced for each user. […] Positive network effects are the main source of value creation and competitive advantage in a platform business.

In the twentieth-century industrial era, giant monopolies were created based on supply economies of scale. These are driven by production efficiencies, which reduce the unit cost of creating a product or service as the quantities produced increase. […] Demand economies of scale are driven by efficiencies in social networks, demand aggregation, app development, and other phenomena that make bigger networks more valuable to their users.

Scaling a network requires that both sides of the market grow proportionally.

The growth of a network can produce negative network effects that drive away participants. […] [W]hen the growth in numbers that enables more matches between producers and consumers also leads to increasing difficulty, or impossibility, in finding the best match. To avoid this dilemma, frictionless entry must be balanced through effective curation.

  • Network effects: a long-lasting network of users; lock-in;
  • Different from price effects and brand effects and virality.
    • price effects: evanescent. “Freemium models create freeloaders”
    • brand effects: sticker but difficult to sustain, and can be extremely expensive
    • “Virality is about attracting people who are off the platform and enticing them to join it, while network effects are about increasing value among people on-platform” “[N]etwork effects keep [people] there.”
  • Improve network effects with
    • frictionless entry
    • curation
      • “the larger your network grows, the better your curation can become—a phenomenon we refer to as data-driven network effects”
      • “poor curation leads to greater noise”


  • Whereas giant industrial-era firms were made possible by supply economies of scale, today’s giants are made possible by demand economies of scale—expressed as network effects.
  • Network effects are not the same as price effects, brand effects, or other familiar growth-building tools.
  • Frictionless entry and other features of scalability maximize the value-building impact of network effects.
  • A two-sided market (with both producers and consumers) gives rise to four kinds of network effects: same-side effects (positive and negative) and cross-side effects (positive and negative). A growing platform business must manage all four.
  • The key to minimizing most negative network effects is quality curation, which increases the chances of a happy match between producer and consumer.
More quotes from chapter 2

Where network effects are present, industries operate by different rules.

As drivers sign on [to Uber] and coverage density rises within a city, a number of striking growth dynamics are set in motion. Riders tell their friends about the service, some even start driving themselves in their spare time. Wait time falls for riders, and downtime falls for drivers. Less downtime means that a driver can make the same amount of money even if fares are lower, because he has more riders during the same number of working hours. Thus less downtime means that Uber can cut fares and stimulate even more demand, which creates a virtuous cycle that increases coverage density still further. […] If Uber attracts too many drivers relative to the number of riders, driver downtimes will go up; if Uber attracts too many riders relative to drivers, rider wait times will go u.

In the case of Uber two sides of the market are involved: riders attract drivers, and drivers attract riders.

[P]latform businesses will often spend money to attract participants to one side of the market. […] [T]he profits to be earned in Market B must outweigh the losses incurred in Market A.

Curious about what separated the successful companies from those that failed [in the dotcom bust of 2000], we examined dozens of cases and found that the failures mostly relied on price or brand effects. By contrast, the successes hit on an idea that really worked—driving traffic from one user group in order to drive profits from another user group.

Threadless is a T-shirt company [whose] business model involves holding weekly design contests open to outside participants, printing only T-shirts with the most popular designs, and selling them to their large and growing customer base. Threadless doesn’t need to hire artistic talent, since skilled designers compete for prizes and prestige. It doesn’t need to do marketing, since eager designers contact their friends to solicit votes and sales. It doesn’t need to forecast sales, since voting customers have already announced what numbers they will buy.

Chatroulette had no registration requirement and no controls of any kind, leading to what became known as the Naked Hairy Men problem.

Firms such as Airbnb, Uber, Dropbox, Thredless, Upwork, Google, and Facebook are not valuable because of their cost structures: the capital they employ, the machinery they run, or the human resources they command. They are valuable because of the communities that participate in their platforms.

Chapter 3 — Architecture: Basic Principles for Designing Successful Platforms

Recap: Challenges of the platform:

  • How do we build a platform that invites participation and creates significant value for all its users?
  • How do we provide tools and services that make it easy for producers and consumers to interact in mutually rewarding ways?
  • How do we design a technological infrastructure that is capable of scaling rapidly, and encouraging positive network effects while minimizing negative ones?

“The natural tendency of those charged with creating a new platform business is to study similar implementations and imitate them. But because no two markets are identical, this strategy often fails.”

Producers and consumers exchange three things: information, goods or services, and some form of currency

Exchange of information

  • Every platform interaction starts with the exchange of information
  • Thus, every platform business must be designed to facilitate the exchange of information
  • Examples:
    • Uber provides information about driver availability and location in response to passenger requests
    • Yelp provides information about restaurants to enable users to choose a place to eat
    • Upwork allows companies and freelancers to exchange information about themselves to facilitate hiring decisions
  • The exchange of information takes place through the platform itself. This is a fundamental characteristic of a platform business.

Exchange of goods or services

  • Following the exchange of information, participants may decide to exchange goods or services
  • Each item exchanged among platform users is called a value unit
  • Goods or services can be exchanged through the platform, or outside of the platform
    • Examples of goods or services exchanged through the platform
      • Facebook, where the value unit is photos, links, and posts
      • YouTube, where the value unit is videos
    • Examples of goods or services exchanged outside of platform
      • Uber, where the value unit is journeys in real cars
      • Yelp, where the the value unit is physical meals consumed in actual restaurants

Exchange of currency

  • When goods or services are exchanged on platforms, they are typically paid for using some for of currency
  • In many cases, this is traditional currency transmitted using credit carts, paypal, or bitcoin
  • There are also other forms of value in which consumers “pay” producers:
    • YouTube, Twitter: Viewers/Followers pay the producer with attention
    • TripAdvisor/Dribble: Community members pay by enhancing the reputation of the producer
  • Attention, fame, influence, reputations, and other intangible forms of value can play the role of “currency” on a platform

The core interaction: The Why of Platform design

The core interaction:

  • the single most important form of activity that takes place on the platform
  • exchange of value that attracts users to the platform in the first place
  • needs to be as easy, attractive, and valuable to users as possible
  • fundamental purpose of the platform is to facilitate the core interaction
  • Three key components: the participants, the value unit, the filter

Platforms should start with one primary interaction. Over time, platforms can enable multiple interactions, but each needs to be designed to meet a particular platform goal and to help users create a new form of value.

The participants

  1. The producer, who creates value
  2. The consumer, who consumes value

Participants may play different roles. Examples:

  • Airbnb: The same person can be both a guest and a host
  • YouTube: Users may upload vides as well as view them

A well designed platform makes it easy for users to move between roles.

The value unit In virtually every case the core interaction starts with the creation of a value unit by the producer. Examples of value units:

  • eBay and Airbnb: the product/apartment listing
  • Kickstarter: the project details
  • YouTube: videos
  • Twitter: tweets
  • Uber: available car categories
  • LinkedIn: profiles of professionals

In each case users are provided with a basis for deciding whether or not they want to proceed with the exchange.

The filter

  • The value unit is delivered to selected consumers based on filters
  • It’s an algorithm to enable the exchange of appropriate value units between users
  • A well designed filter ensures platform uses receive only value units that are relevant and valuable to them
  • A poorly designed filter (or no filter) means users may be flooded with units they find irrelevant and valueless, which may drive them to abandon the platform

Every platform makes use of filters to manage the exchange of information:

  • Uber: Potential riders share their location and requested time, based on which the most relevant available cars can be filtered and presented to the user
  • YouTube: Past videos watched, interests and channels that the user subscribed to make up parameters that the YouTube algorithm uses to match most relevant and valuable videos

Participants + Value Unit + Filter → Core Interaction

“Successful platforms begin with a single core interaction that consistently generates high value for users”

“A valuable core interaction that is easy, even enjoyable, to engage in attracts participants and makes the emergence of positive network effects possible”

Platforms don’t create value units, but:

  • They provide the infrastructure within which value units are produced
  • They can foster a culture of quality control (of value units)
  • They develop filters that are designed to deliver valuable units while blocking others

Pull, Facilitate, Match: The How of Platform design

The purpose of a platform is to make core interactions inevitable my making them valuable to all participants. This section talks about how to do that.

  • The platform must pull both producers and consumers to the platform to enable their interaction.
  • It must facilitate their interactions by providing them with tools and rules that encourage valuable exchange.
  • It must match producers and consumers effectively to connect them in a way they find mutually rewarding

All three functions must be performed well for a platform to succeed.

  • If a platform fails to pull participants in, there will be no network effects.
  • If it fails to facilitate interactions, participants will be discouraged and alienated.
  • If it fails to match participants, their time and energy will be wasted, driving them away.

Pull Chicken or egg problem

  • Users won’t come to a platform unless it has value and a platform won’t have value until users use it.
  • Most platforms fail because they never overcome this problem

Retention problem

  • Platforms need to keep the interest of users who signed up to the platform
  • Facebook discovered that users found the platform valuable only after they had connected to a minimum number of other users. Facebook shifted its marketing efforts away from recruiting new members to helping them form connections.

Feedback loop

  • A user’s action generates a response that causes the user to re-engage
  • Constant stream of self-reinforcing activity
  • Can help swell the network, increase value creation, and enhance network effects
  • Can involve just one user or multiple
    • Example for a multi-user feedback loop is the Facebook news feed

Facilitate Process of facilitators refers to the platform-owned infrastructure as part of which value can be created and exchanged.

It’s important to make it as easy as possible for producers to create and exchange valuable goods and services.

Facilitating interactions may involve reducing barriers to usage.

  • Instagram enabled users to snap, modify, and share pictures in three clicks on a single device

Sometimes increasing barriers to usage can have positive effects, leading to more high quality participants. Examples include ID verification on Airbnb, or background checks for a babysitting platform.

Match Successful platform efficiently match the right users with one another to ensure that the most relevant goods and services are exchanged.

Platforms match users using the available data about:

  • producers
  • consumers
  • the created value units
  • the goods and services to be exchanged

“The more data the platform has to work with — and the better designed the algorithms used to collect, organise, sort, parse, and interpret the data — the more accurate the filters, the more relevant and useful the information exchanged, and the more rewarding the ultimate match between producer and consumer.

Successful platforms create mutually rewarding matches on a consistent basis.

  • Continual improvement of data acquisition and analysis is important

All three functions (pull, facilitate, match) are essential to a successful platform. However, it’s possible for a platform to survive, at least for some time, thanks to its strength at one function. Examples:

  • Craigslist is poor at facilitate and match but unmatched at pull
  • YouTube and Vimeo can co-exist because YouTube is excellent at pull and match, whereas Vimeo focuses on facilitate

Beyond the core interaction

As platforms mature, they may successfully shift their efforts on peripheral aspects of the platform that only have an indirect relationship to the core interaction or add entirely new interactions.

  • Uber started acting as a guarantor for car loans so that potential drivers without a credit history could get cars and become producers on the platform.

  • Linkedin added additional interactions which all give users additional reason to visit the platform

    • Groups where discussions can be conducted
    • Ability for recruiters to contact candidates
    • Ability for users to thought lead using posts

Applying the end to end principle to platform design

Platforms need to find a balance between two competing forces:

  1. Innovation can lead to excessive complexity, which makes the platform difficult to navigate.
  2. A platform that fails to evolve by adding desirable new features is likely to be abandoned by users who discover a competing platform with more to offer.

The end to end principle can help find this balance:

“Activities that are not central to the workings of the network but valuable only to particular users should be located at the edges of the network rather than at its heart.“

“Application specific features should reside in the layer of process at the edge or on top of the platform, rather than at roots deep within the platform. Only the highest-volume, highest-value features that cut across apps should become part of the core platform.”

A well designed platform consists of a stable core layer that restricts variety, sitting underneath an evolving layer that enables variety.

The power of modularity

  • In the long run, a successful platform must have a modular approach
  • Subsystems can be designed independently as long as they adhere to overall design rules and connect to the rest of the system only through standard interfaces
  • Modularity can be achieved with APIs enabling outside access to core resources
  • Not publishing APIs (in favour of for example speed) makes it much harder to mobilise an external ecosystem of developers who can build on top of the core platform

Re-architecting the platform

It’s possible to re-architect platforms in order to make them more modular. One of the most successful such changes was the standardisation of hardware interfaces (PCI and USB) by Intel in the 1990s.

Iterative improvement: The Anti-design principle

  • Platforms cannot be entirely planned, because users ultimately decide how to use them.
  • As such platform designers should leave room for serendipitous discoveries
  • Platforms should closely monitor user behaviour which will reveal unexpected patterns

The best platform allow room for user quirks, and they are open enough to gradually incorporate such quirks into the design of the platform

Takeaways of Chapter 3

  1. The design of a platform should begin with its core interaction
  2. Three key elements define the core interaction:
  3. the participants,
  4. the value unit,
  5. and the filter. Of these, the value unit is the most crucial, and often the most difficult to control
  6. To make the core interaction easy and inevitable, a platform must perform three crucial functions:
  7. pull,
  8. facilitate,
  9. and match.
  10. Platforms can grow beyond the core interaction.
  11. New kinds of interactions may be layered on top of the core interaction
  12. Platforms need to be designed
  13. to make mutually satisfying interactions easy for large numbers of users
  14. and to leave room for serendipity and the unexpected

Chapter 4 - Disruption: How Platforms Conquer and Transform Traditional Industries

Platform: A place where producers and consumers can come together in interactions that create value for both parties. Main difference with traditional platform businesses like bazaars and stock markets is the addition of digital technology. Technology improves: reach, speed, convenience, and efficiency. Uber is a winning deal for everyone involved, the only losing party is the taxi driver who lost a job.

A Capsule History of Digital Disruption

Software is eating the world. — Marc Andreessen Digital disruption has happened in two stages. We’re in the second stage now:

  1. Efficient pipelines ate inefficient pipelines
  • 1990s online systems for distributing goods and services outcompeted incumbent industries.
  • Low marginal costs of distribution, target and serve much larger markets with smaller investments.
  • Newspapers and classifieds took a hit as did retail and mail order shopping, dvd rental and music.
  1. How platforms eat pipelines
  • “Software is eating the world” -> update to “Platforms are eating the world”
  • Uber, Airbnb, Upwork, Amazon, Apple, Google — All growing businesses based on platforms.

How and why is this happening? Internet is no longer just a distribution channel (pipeline). It’s a creation infrastructure and coordination mechanism that can be used to create entirely new business models. Two significant advantages over pipelines:

  1. Superior marginal economics of production and distribution
  2. Traditional businesses, like hotels need large capital investments and staff to expand, whereas a platform business like Airbnb has a near-zero marginal cost to adding new rooms to the platform.
  3. Network effects
  4. Network effects make it possible to build ecosystems with millions of remote participants, much larger than a traditional pipeline business. Positive network effects serve as a flywheel, the more activity the more growth. It’s obvious on a platform like Upwork: more freelancers attract more businesses and more businesses attract more freelancers.

The Impacts of Platform Disruption on Value Creation, Value Consumption, and Quality Control

Platforms have economic advantages over pipeline businesses allowing them to grow faster. By reconfiguring the business processes of value creation, value consumption, and quality control platform dynamics can be improved. Reconfiguring value creation to tap new sources of supply The goal is to minimize friction for the creators of supply in your platform, some examples:

  • Video platform Viki made it easy to subtitle videos in Korean, enthusiasts created more than any company could have ever done alone.
  • Facebook did a similar thing buy making it easy to source translations of it’s website.
  • Airbnb runs events for hosts to teach best practices
  • Airbnb sent professional photographers to make listing easier
  • Uber has sign up bonuses and leasing programs for drivers to get started more easily
  • Amazon Kindle offered quick and easy access to large readership

Reconfiguring value consumption by enabling new forms of consumer behavior Similar objective, however this time focused on comsumers. By changing behavior of consumers we can increase usage of products and services in ways that would have been unimaginable a few years ago.

We are hopping into strangers’ cars (Lyft, Sidecar, Uber), welcoming them into our spare rooms (Airbnb), dropping our dogs off at their houses (DogVacay, Rover), and eating food in their dining rooms (Feastly). We are letting them rent our cars (RelayRides, Getaround), our boats (Boatbound), our houses (HomeAway), and our power tools (Zilok). We are entrusting complete strangers with our most valuable possessions, our personal experiences—and our very lives. In the process, we are entering a new era of Internet-enabled intimacy.

Reconfiguring quality control through community driven curation Curation is valuable to keeping participants happy. Without it, quality will fall and you can lose traction. With scale, curation becomes more of a challenge. Lean on your community and data. New platforms are often criticized because they fail to offer quality and reliability of traditional competitors:

  • YouTube - indecent and pirated content
  • Airbnb - raids on apartment parties
  • Wikipedia - pages positioning falsehoods as fact

In the early days of a platform negative network effects from increased supply lead to degraded quality. Strong curation encourages desirable behavior and can spark growth of the platform that far exceeds the growth potential of a pipeline based business. It’s important to make sure curation mechanism doesn’t break down as the platform scales. Successful platforms use their data to replace manual curation with automation

Structural Impacts of Platform Disruption

De-linking assets from value In many cases value can be derived from ownership of physical asset, but it may not be put to it’s highest and best use due to underutilization or a mis-match between the owner’s business and the best use of the asset. By removing the connection between ownership and use of a productive asset a platform business can increase efficiency of use and overall value.

  • NY State - Designed a smart market for energy. Small energy sellers, sell to large buyers, who then redistribute to consumers. By coordinating supply and demand across power generation and storage infrastructure across the state higher efficiency and value is realized for all.
  • Health care - Many hospitals run expensive MRI machines at 40-50% utilization. By separating ownership from usage they’re able to time share the device with small clinics who cannot afford their own machines.
  • Farmers - In Australia, water use reform led to a system where water rights are separated from property ownership. A private company created a platform for trading water. A farmer of a low-value crop then was able to stop farming, trade the water and it was put to better use.

Re-intermediation With the rise of the internet disintermediation was expected. Think removing the travel agent from the booking of airline tickets. In reality, platforms are re-intermediating markets by introducing new types of middlemen rather than only eliminating them. Typically re-intermediation involves replacing non-scalable and inefficient agents with online, often automated tools and systems. Networked platforms are more efficient because of their ability to scale. Platforms rely on algorithms and data rather than manual interactions.

  • Music A&R - Rather than working with labels, they’re more independent and scouting talent online via platforms like YouTube and Soundcloud.
  • Literary agents - Search for new authors on Quora and Medium.
  • Advertising - Small businesses run campaigns with AdWords without relying on an agency.
  • Social feedback - Yelp, Angie’s List, TripAdvisor are part of a new industry that certifies the quality of a product or service.
  • New winners and losers - In traditional book industry publishers absorb (and spend) most revenues, paying author royalties of 10-15%. On Amazon, authors retain 70% of revenues.

Market aggregation

Market aggregation is the process whereby platforms provide centralized markets to serve widely dispersed individuals and organizations.

Because of fragmentation in a given industry, search costs for a consumer can be very high. By aggregating all of this data a platform has the opportunity to realize value through an improved experience for the consumer. Amazon Marketplace, Alibaba, Etsy, eBay bring haphazard disorganized business into an organized space that is valued by the consumer. See also: Aggregation Theory – Stratechery by Ben Thompson

The Incumbents Fight Back: Pipelines Becoming Platforms

Platform businesses are disrupting traditional businesses. For incumbents to survive, they’ll need to reevaluate their business models. Ask yourself:

  • What processes can be delegated outside the company?
  • How can we empower partners to create new value for our existing customers?
  • Can we network with competitors to create new value for existing customers?
  • How can the value of our products be increased by data, personal connections, and curation?

Pipeline businesses traditionally scale in one of two ways:

  1. Vertical integration - Owning and integrating a greater length of the value-creation-and-delivery pipeline.
  2. Horizontal integration -Expanding by widening the pipeline to push more value through it.


  • Nike bought wearable devices company, FuelBand.
  • Apple connects it’s products and services together. By owning more products they are much more useful.
  • UnderArmour competing with Nike to build a platform business in the wearables market, purchased MapMyFitness, MyFitnessPal, and Endomondo.

Can any product or service become the basis of a platform business? Here’s the test: if the firm can use either information or community to add value to what it sells, then there is potential for creating a viable platform. This creates huge opportunities for a lot of firms.

Chapter 5: Launch - Chicken or Egg? Eight Ways to Launch a Successful Platform

PayPal Case Study

Peter Thiel & Max Levchin launch Confinity

  • Goal was enabling money transfers on Palm Pilots and PDAs w/infrared ports
  • Shut down in 2 years after gaining only 10k signups

Launched PayPal based around an engineer's POC demo of making payments via email

  • Spending $10 million per month on the business
  • had to solve chicken or egg problem

2-sided Markets where both sides are important: Which comes first?

  • Have to have sellers that would accept the new form of payment or buyers wouldn't use it
  • Have to have buyers that want to use it or sellers won't invest time/effort/money to accept it

PayPal solved problem through series of ingenious strategies

  • Reduced friction involved in accepting online payments (just need email address and credit card)
  • Gave new customers $10 for signing up and $10 for each friend that signed up
  • Focused on user commitment over user acquisition (people had to use product to spend their free $10)
  • Ebay bot would buy item and insist on paying using PayPal

Benefited from virtuous feedback loop

  • Users saw convenience of paying with PayPal
  • Insisted on only paying with PayPal
  • More sellers wanted to offer it as a way to pay (would even add PayPal logo to their site)
  • Logos informed more buyers of existence of PayPal

Grew from 100,000 users to 1,000,000 in 3 months

Ebay initially built alternative to PayPal, but failed to gain traction

  • Bought PayPal in 2002 for $1.4 billion in stock

Heart of Platform Marketing: Designing for Viral Growth

Focus on pull strategies

  • customers have access to nearly unlimited marketing/communication channels
  • must craft messages that pull customers toward your products
  • marketing must be baked into the platform (incentivize user sharing)

The Incumbents' Advantage: Reality or Illusion?

Why don't large incumbents always win?

  • They have existing value chains, partnerships, pool of talent, & loyal customer base
  • They become complacent because they expect they'll have time to watch a slow rise of a challenger
  • Network effects can change the market quickly so encumbents might be left behind

There are Many Ways to Launch a Platform

Strategies that work for one platform won't work for another

  • YouTube:

    • first democratic upload platform
    • focused on incentivizing content creators
    • allowed off-site embedding of videos to spread word
    • shared ad revenue with major creators
  • Megaupload:

    • Late-mover problem, producers had no reason to leave YouTube
    • Fcused on consumers by seeding platform with internally created content
    • Went deep in verticals that were policed on YouTube (like pirated vids and porn)
  • Vimeo:

    • focused on providing high quality tools that attracted creators that felt neglected by YouTube
    • provided better infrastructure for uploading videos

It's important to know the value propositions offered by competitors so you can carve out your niche

Eight Strategies for Beating the Chicken-or-Egg Dilemma

First strategy is to avoid answering the chicken or egg problem at all:

  1. Follow the rabbit strategy
  • Use non-platform demonstration project to model success (to attract both users/producers to new platform)
    • EX: Amazon - operated as successful online retailer to attract customer base, THEN opened up to external producers with Amazon Marketplace

Not always possible to "follow the rabbit", in these cases, you need techniques to attract base of users on both sides of the market:

  • Staging value creation

    • managers create value to attract 1 or more users they can demonstrate the benefits of the platform to
    • Initial users create more value to attract other users
    • Positive feedback loop ensues
  • Designing the platform to attract one set of users

    • Create tools/products/services to attract either consumers or producers
    • Critical mass of users on either side of the marketplace will attract users on the other side
  • Simultaneous on-boarding

    • Create conditions that provide value to users even when network is small
    • stimulate bursts of actiity to attract consumers/producers
    • create larger value to start getting network effects

Strategies for when you have to solve the chicken/egg problem:

  1. The piggyback strategy
  • Connect w/existing user base from different platform
  • Provide value to incentivize those people to use your platform
  • Ex: Scraping Craigslist to create listings on your site
  1. The seeding strategy
  • Create value units that will be relevant to at least one set of users
  • Once they are on the platform, use them to attract users that want to engage with them
  • Ex: Google having $5 million in prizes to people that create best-in-category app for Android
  1. The marquee strategy
  • provide incentives to attract members of a key user set onto your platform
  • focus on users that are so important their participation will make/break your platform
  • Ex: Microsoft bought Bungie and used "Halo" as a platform selling game for the Xbox
  1. The single-side strategy
  • Create business around products/services that benefit a single set of users
  • Later turn into a platform by attracting second set of users who want to engage with the first set
  • Ex: OpenTable created tools for restaurants to manage their reservations then extended it to customers looking for available reservations
  1. The producer evangelism strategy
  • Design platform to attract producers who can induce their customers to become users of platform
  • Ex: Kickstarter targets creators that need funding who then encourage their own followers to join the platform to donate
  1. The big-bang adoption strategy
  • User 1+ push marketing strategies to attract high volume of interest/attention to your platform
  • Result is almost simultaenous on-boarding effect
  • Ex: Twitter spent $11k on giant screens showing live tweets at SXSW which tripled daily tweets from 20k to 60k per day
  1. The micromarket strategy
  • Target a tiny market comprised of members who are already engaging in interactions
  • Platform provides effective matchmaking usually only seen in larger markets
  • Ex: Facebook launching a social network in the closed community of Harvard

Viral Growth: The User-to-User Launch Mechanism

Viral growth is a pull-based process based on encouraging users to spread word about the platform to other potential users

4 elements necessary for viral growth

  • The sender
    • Spread their own creations and indirectly generate awareness of and interest in your platform
    • Important to make spreading as easy as possible for senders
    • Can use inorganic (e.g. financial) incentives to get people to be senders, but this can quickly get very expensive
  • The value unit
    • fundamental unit of virality, important to be spreadable
  • The external network
    • an existing platform you can leverage
    • these platforms will often restrict usage if you become too 'dominant' so you have to be careful
  • The recipient
    • senders target, will respond if they find the value unit valuable
    • will spread value unit further if they like it

Chapter 6: Monetization

Slogan: "Get users first, monetize later"

  • Should you charge users/providers for listing something on your platform?

    • Answer: NO! The value of a platform is in the data. Do not add friction to people looking to join your platform.
    • Details:
      • Entry into the platform should be as easy as possible. Do not charge $$ for entry. (example: 99designs does not charge a fee for listing a project. Also does not charge designers for joining)
      • Do not let your monetization strategy ruin your network effect.
      • Less friction = more data! If anything, platform should provide tools for making it easier to join. It should be desirable to have a profile/presence on the platform.
  • What can a platform charge for?

    • Answer: "They can charge users for the value they accrue from the ecosystem". Charges should be "levied on deal completion, not at the time of the listing"
    • Details:
      • Having fees that are performance based make them "feel negligible" because it is simply "skimming" off a small fraction of the transaction.
      • There are other opportunities for charging for value. Example: if a business is failing to gain customers on the platform. Could charge a fee for a listing review. That helps them improve their ability to gain customers.

Example of failure

  • Zvents was an event listing platform (all local events in 1 place). They couldn't charge the event organizers because that would result in less events (then consumers would stop visiting).
  • They attempted charging for promoted listings, but this was not enough revenue.
  • Eventually failed and was acquired. Platform is now used by Ebay to direct ticket sales to StubHub (end result: platform still free, links to StubHub to generate revenue)

Example of success

  • Struggled finding correct monetization strategy. Their solution was interesting. They had a problem of too many low quality meetups on their site. Solution: started charging $19/month to each organizer. HUGE OUTRAGE from the community. But it worked. Meetup quality improved, platform thrived. Tools were good enough for organizers that they were willing to pay.

TLDR: They succeeded because charging $$ increased the quality of their network.

Questions to ask yourself when creating $$ strategy

  1. "How can we generate revenue without reducing positive network effects?" (Zvents couldn't, so they failed)
  2. "Can we devise a pricing strategy that strengthens our positive network effects while reducing our negative network effects?" (think meetup example)
  3. "Can we create a strategy that encourages desireable interactions and discourages undesirable ones?"

Common ways to monetize a platform

1. Transaction fee

Charge a fee on completion of a service

  • Uber
  • Upwork
  • AirBNB
  • GitHub Marketplace

Problem: Buyers and sellers will meet on the platform and they do business outside to avoid the fee.

Solution: The platform needs to provide so much value that it is preferable to do the transaction through it. Example: Clarity connects people for quick consulting calls. Their platform allows for easy billing by the minute. This makes it preferable for consumers.

Another example: Upwork and task rabbit make it difficult for consumers/providers to communicate outside of the platform. Making it less likely they will do a deal. They add RATINGS so that providers want the deal to happen through the platform. Higher ratings = higher sales. Only people who buy through platform are allowed to rate.

2. Charging for access

Charge producers for access to a community

  • Dribbble
  • LinkedIn
  • Stack Overflow

^ Each has a highly engaged community. Providers pay money to post job listings that are seen by the community.

Success: This works very well for Dribbble. The community is incentivized to put up their best work (to attract employers). Then employers have easy access to portfolios for evaluation.

3. Charging for enhanced access
  • Yelp
  • Dating websites

Restaurants can pay for higher placement on Yelp. Dating websites charge for more profile views / ability to see more people.

Problem: this needs to be done with care because it can cause a negative network effect.

Key: When doing this, be sure to make it clear that enhanced access is being paid for. "X business is at the top of the list because they paid"

4. Charging for enhanced curation

When the quantity of content on a platform becomes too great. Opportunity to charge for curation.

Example: Baby sitting platform. Charges parents money for access to their list of baby sitters (platform screens babysitters to ensure quality).

Chapter 7: Openness

🔖 Definition of openness:

A platform is 'open' to the extent that:

  1. No restrictions are placed on participation in its development, commercialization, or use; or
  2. any restrictions - for example, requirements to conform with technical standards or pay licensing fees - are reasonable and non-discriminatory, that is, they are applied uniformly to all potential platform participants.
  • Openness levels exist on a spectrum, and where a business decides to land on that spectrum affects usage, developer participation, monetization, and regulation.
  • There are pros/cons at ever point in the spectrum
  • Defining who has access, and how they can participate is very complex. Should be at the top of a platform managers agenda at all times.

💼 Wikipedia:

  • Wikipedia is the most popular reference site in the world
  • Is a universally accessible, unlimited source of data that should be reliable
  • There are a lot of cases of misinformation on the site. Example: people editing the "Murder of Meredith Kercher" 8,000 times, mostly by people who are biased towards believing that Amanda Knox is guilty.
  • Wikipedia is an incredibly open platform, and this is a good example of the tradeoffs inherent in providing something that is openly accessible while trying to still maintain high quality. If Wikipedia were closed, it would reduce participation.

💼 Apple's transition from closed to something more balanced:

  • in 1980s, Jobs chose to keep the Apple Macintosh a closed system, while Microsoft opened an OS to outside developers and licensed it to many computer manufacturers.
  • Microsoft had a flood of innovation, Windows claimed a much larger share of the personal computer market.
  • In 2000s, Jobs opened the iPhone OS, made iTunes available on Windows. This was a better balance, Apple captured more of the smartphone market compared to Nokia and Blackberry.
  • Jobs interpreted the openness spectrum as "fragmented" (open) vs. "integrated" (closed); this terminology biased the debate to favor closed, controlled systems.

💼 Myspace vs. Facebook:

  • Myspace dominated social networking before Facebook. It released a ton of features, but constrained engineering resources meant they were often buggy and had poor user experience. Because Myspace kept their site closed to outside developers, the problem continued.
  • In retrospect, cofounder Chris DeWolfe says that they should have focused on 5 - 10 key features and let the rest be built by outside developers.
  • Facebook was initially closed to outside innovators, but opened to dot-com users in 2006 and started gaining on Myspace.
  • The huge shift came from the launch of Facebook Platform, which launched in May 2007 to help developers create apps. By November, there were 7000 outside applications. Myspace reacted by also opening to developers in 2008, but it was too late.
  • Myspace's self-serve advertising feature opened the platform to significant amounts of inappropriate content, further degraded Myspace's attractiveness to users.
  • This is an example of a platform being simultaneously too closed and too open.

💼 Youtube:

  • Highly open system = delivers a wide array of content from all types of content creators; is a true platform vs a distribution system.

🔑 Three kinds of openness decisions that managers face:

  1. Manager/sponsor participation
    • Two entities behind a platform: a firm that manages the platform, and the firm that sponsors the platform
    • Sometimes these are the same thing: Facebook, Uber, eBay, Airbnb.
    • Sometimes these are not the same: Manager has more influence over daily operations and the developers, the sponsor has legal and economic control, as well as long-term strategy.
    • See next 🔑 for examples
  2. Developer participation
    • Managers need to control the level of openness to developers
    • See section on "🔑 developer participation" below.
  3. User participation
    • Managers need to control the level of openness to users
    • See section on "🔑 user participation" below.

🔑 1. Management and sponsorship of a platform can be controlled by:

  • Depending on 4 combinations, there are different patterns of openness and control, different advantages and disadvantages. A platform can also migrate from one model to another.

  • The Proprietary Model: A single firm does it all (provides the greatest control and facilitates closed system of operation.)

    • Underlying hardware/software for the Mac OS and mobile iOS are controlled by Apple.
    • Feels like the dream; capture an entire market through developing a new technology standard and maintaining control over it. But this can backfire:
      • 💼 Betamax videotape sponsored by Sony vs. VHS sponsored by JVC.
        • Betamax was higher quality, but this "war" was decided based on differing sponsorship/management strategies.
        • Sony used proprietary model, JVC used licensing model.
        • Manufacturers could produce VHS recorders and players, led to more volume and falling prices, more consumers buying. Movie studios issued more products in VHS format.
        • VHS won out, although JVC didn't get much financially in the long run.
  • The Licensing Model: A group of firms manages, one firm sponsors (open at one end, closed at another)

    • Google sponsors the Android OS, but encourages other firms to supply devices.
  • The Joint Venture Model: Single firm manages, group of firms sponsor (open at one end, closed at another)

    • Orbitz launched as a joint venture with many major airlines.
    • Can be challenging to make key decisions when there's a committee of owners with different goals.
      • 💼 Visa, which was sponsored by a number of banks since the 70s, ended up shifting the business to a self-contained entity that fit into the proprietary model in 2007.
  • The Shared Model: Groups of firms manages, another group of firms sponsors (open to both managers and sponsors)

    • Linux OS has a platform connecting app developers to consumers. Managing companies include TiVo, Roomba, Ubuntu, Qualcomm, while corporate sponsors include IBM, Intel, HP, Fujitsu, NEC, Oracle, Samsung.
  • How do you choose which sponsorship/management model to choose?

    • Tradeoffs inherent in these models are complicated.
      • 💼 Apple vs. Google smartphones:
        • The iPhone (more controlled) considered more user friendly and aesthetically pleasing than Google's smartphone (less controlled).
        • Android Open Source Platform (AOSP) permits experimentation by any company.
        • By 2014, open innovation led Android to 80 percent market share, Apple at 15 percent.
        • Does this mean Google is better? Maybe not. AOSP OS doesn't channel users to Google's online services, which means Google gets no revenue or data flow from AOSP devices. Google is now reversing course and closing Android.
    • Choose based on why the platform is being developed and the goals of those designing it:
      • 💼 RFID tech
        • Used to create smart tags for products for inventory control.
        • Uses a shared model: sponsored by many retailers, manufactured by many companies.
        • RFID doesn't generate huge profits, but it reaches the goal of the sponsors: make the technology simple, accessible, and affordable.

🔑 2. Developer Participation

  • Three kinds of developers:
    1. Core developers: create the core platform conditions that provide value to platform participants.

      • Goal: get the platform to users, deliver value through tools and rules around the core interaction. Responsible for basic platform capabilities.
      • Generally employed at the platform management company.
    2. Extension developers: Add features and value to the platform and enhance its functionality.

      • Usually outside parties. Managers have to decide how open they want to be to these developers and usually find ways to consequently profit from the value that this group creates.
      • Ex. People who produce apps sold via the iTunes store
      • Companies will usually create an application programming interface (API) to regulate and encourage these developers in their participation.
      • 🔖 An API is a standardized set of routines, protocols, and tools for building software applications that makes it easy for an outside programmer to write code that will connect with the platform infrastructure.
      • 💼 Walmart vs. Amazon's APIs: Amazon has 300 API "mashups", Walmart has one API, which is partially why Amazon pulled ahead of Walmart in the stock market in 2015.
      • 💼 Salesforce generates 50 percent of its revenues through APIs.
    3. Data aggregators

      • Enhance the matching function of the platform by adding data from multiple sources. They take data from the platform and sell to other companies for activities such as ad placement, and share profits with the original platform company.
      • When it works well, it provides good matches for consumers. When it doesn't, it's very creepy.
      • "Managing [this] appropriately poses an enormous ethical, legal, and business challenge."

      💼 Developer Participation Case Study: Airbnb

      • Core developers:
        • provides infrastructure that allows guests and hosts to interact.
        • manages background functions that reduce transaction costs (ex. default insurance contracts for both parties)
      • Extension developers:
        • Ex. Developer now offers professional support under "Airbnb photography service" to create compelling images for hosts.
        • If Airbnb is too closed, makes it hard for extension developers to provide value. If too open, increases existence of low-quality providers which damage Airbnb's reputation, too many providers competing for the same profits.
        • Airbnb takes the middle route: it has an API, but it is not open to all developers.

      💼 Developer participation and APIs case study: The Guardian

      • The Guardian is a British daily newspaper. Originally closed to extension developers, but decided to "open in" the website "by bringing in more data and applications from the outside", and "open out" the site, by "enabling partners to create products using Guardian content and services on other digital platforms".
      • For "open out", they created a set of APIs with three different levels of access:
          1. Keyless, the lowest access tier: anyone can use headlines, metadata, and information architecture without permission / without needing to share revenue.
          1. Approved, second access tier: allows time and usage restrictions, ad revenues are shared.
          1. Bespoke, highest tier: support package and gives unlimited content for a fee.
        • This has led to more products like a Politics API, and an app framework. Over 2,000 extension developers signed up in the first year. APIs can attract extension developers.

      When does the outside developer threaten that of the platform, and how should managers respond?

      • Three principles to use when assessing the situation:
          1. Based on the value created by a specific extension app. Don't let an outside developer control a main source of user value. One way to address this is to acquire the app.
          • 💼 Apple owns most of the main apps on the iPhone. Bought SRI, the company behind Siri.
          1. If an app has potential to become a powerful platform itself, the manager should try to own it or replace it.
          • 💼 Google Maps was a popular feature on the iPhone, very powerful and popular. Apple decided to make its own mapping software, not dependent on Google anymore.
          1. If specific functionality is reinvented by many extension developers and platform users like it, the manager of the platform should acquire this functionality and make it available through an API. Ex. Video/audio playback, photo editing, voice commands. = more innovation and better service for users.

🔑 3. User Participation

  • Managers need to address 🔖 producer openness: the right to freely add content to the platform
  • Goal is usually high quality content, which means platforms cannot generally be completely open.
  • 🔖 Curation: content protection process that can be fine-tuned to land on the right level of producer openness. Usually takes the form of screening and feedback at critical points.
    • Options: Can use human moderators, although this can be time-consuming and costly. Alternative is using the users themselves via software tools, although this is hard to design and implement.
    • When curation tools bias towards openness, it can let in offensive and dangerous content. When they bias towards restrictions, valuable content can be lost. Platform managers need to devote a lot of resources, including human "eyeballs"/human judgement, to getting better at this balance.
    • 💼 Wikipedia: originally tried to be completely open, and assumed that users would be able to curate content/challenge biases. But reality was messy.
      • Now relies on community standards and social pressure.
      • When that's not enough, uses other methods like VandalProof (software that highlights articles editing by "bad actor" users), tagging tools, blocking/prevention systems.

🔑 Two separate platforms may compete on the basis of how they differ in level of open/closed

  • There are a lot of factors in being competitive with other platforms, which can result in a never-ending balancing act and continuous adjustements of a platform's level of openness.

🔑 As platforms mature, they tend to grow in the direction of openness (although sometimes the opposite direction)

  • Changes in open/closed come from reevaluation and adjustment of curation processes.
  • Goal: great high quality platform content and service value.
  • The challenge is finding a balance as openness policies evolve.
  • Example of struggle for control:
    • SAP, a German company that makes software for large enterprises to manage internal operations. Partnered with ADP for payroll processing, which gave ADP the opportunity to replace SAP as the primary manager of the customer relationship.

Chapter 8: Governance

🔖 Governance: "the set of rules concerning who gets to participate in an ecosystem, how to divide the value, and how to resolve conflicts." The goal is to create wealth that is fairly distributed amongst those who add value to a platform.

🔑 Three fundamental rules of good governance

  • Always create value for the consumers you serve;
  • Don't use your power to change the rules in your favor;
  • Don't take more than a fair share of the wealth.

💼 Keurig

  • Brian P. Kelley, CEO of Keurig Green Mountain Coffee Company. Keurig 1.0 had become ubiquitous. It's a one-sided platform.
  • In 2012, patents for the Keurig coffee pod expired. Competing companies began selling pods that works in Keurig machines and were cheaper. (Similar to extension developers)
  • Early 2015, launched Keurig 2.0, a next-gen coffee brewer. Included a scanning device to prevent any non-Keurig pods from working in the machine. Consumers were angry, vocal. Sales fell 12 percent.
  • This was a failure of ecosystem governance. Could have opted to become a beverage ecosystem with different options, suppliers, and services, but instead excluded suppliers that its customers liked which eliminated variety and freedom. It placed its interests above all others.

🔑 Governance is even harder when a platform is multi-sided

  • Involves many interests that don't always align. Ex. LinkedIn making developers mad when they turned off access to APIs, Twitter doing a terrible job of dealing with abuse.

🔑 Large platforms effectively regulate lives

  • Large platform businesses are like nation-states. With user bases that eclipse national populations and immense economic impact, giants like Facebook, Alibaba, and Google effectively regulate millions of lives.
  • Platforms can learn from cities and states which have evolved the system of good governance over thousands of years, as they are faced with the same challenge of creating wealth and distributing it fairly.
  • Evidence suggests that "just governance" is crucial in the ability of a state to create wealth, even more so than natural resources, good agricultural conditions, etc.
  • Absolutely free markets cannot be relied on to produce results that are fair to all involved.
    • 💼 Singapore
      • Before 1959, Singapore had almost no natural resources. Relied on Malaya (now Malaysia) for defense and clean water. Lots of corruption. Per capita GDP below $430.
      • Lee Kuan Yew became prime minister in 1959. He changed the system of governance, introduced the British system of justice and rule by Law. Addressed corruption (made illegal activities less attractive by raising salaries of civil servants). Enforced rules. Open government was encouraged by creation of multi-cultural councils giving voice to religious and ethnic groups.
      • Singapore now has a government that ranks with New Zealand and Scandinavian countries, one of the least corrupt in the world.
      • By 2015, per capita GDP = $55,182 (higher than the U.S.), annual growth was 6.69 percent.

🔑 Market Failures

  • 🔖 Market failure: when "good" (fair) interactions fail to occur, or "bad" interactions do. 4 main causes:
      1. Information asymmetry
      • When one party knows facts that other parties don't, and uses that for personal advantage.
      1. Externalities
      • When spillover costs or benefits accrue to anyone not involved in a given interaction.
      • Negative externality example: Friend gives away your private contact information to a gaming company for some points.
      • Positive externalitities are a bit more ambiguous. 💼 Netflix analyzes the behavior of someone with similar tastes as you, and that ends up giving you better movie recommendations. Users in this system are less likely to complain. From business standpoint, this is still a problem "since they reflect value that is not being fully captured by the platform".
      • Ideally, based on economic theory, every value would be accounted for and credited accurately.
      • Similar to a public good, where the value is not fully captured by the party that created it. Individuals tend not to create these unless a governing mechanism is put in place to recognize and reward the effort.
      1. Monopoly power
      • When one supplier in an ecosystem becomes too powerful (through control of the supply), and uses the power to demand higher prices/special favors.
      • 💼 Zynga, games company, in 2009/2010, very powerful on Facebook. Led to conflicts around sharing user information, revenues, etc.
      1. Risk
      • The possibility that something unexpected and essentially unpredictable may go wrong, turning a good interaction into a bad one.
  • 💼 eBay
    • Most interactions are basically fair, even when there is a "winner" and "loser". Sometimes, manipulative or deceptive actions take place.
    • Example of a market failure: A group of sellers noticed that some eBay inexperienced sellers would misidentify goods (often misspelling the brand name), and noticed that they could scoop up valuable items at a lesser price, and resell on eBay for a profit. "Allsop's Arctic Ale" was from a crew that navigated the Arctic in the 1850s, the only two bottles left were eagerly sought after — ultimately was sold for $304 due to the misspelled "Allsopp" in the listing, then purchased and re-sold for $78,100.

🔑 Tools for platform governance

  • Four main sets of tools, generally need to use all of these:
      1. Laws
      1. Norms
      1. Architecture
      1. Markets
  • Need all of these tools to change social behavior.
    • 💼 Reducing the harmful effects of smoking, used all four tools: laws (ban cigarette sales to minors, no smoking in public spaces), norms (social pressure or advertising to stigmatize smoking), architecture (air filters that clean the air, smokeless devices), and market mechanisms (taxing tobacco products or subsidizing "quit smoking" programs).

🔑 Tool 1: Laws

  • Laws can be tricky. Laws can apply to platform businesses and participants, but it's not always clear who is responsible and should be held accountable.
  • Case law generally doesn't hold platforms accountable for misdeeds of users, although platform owners actually can regulate and control the behavior of users.
  • Governance within the platform business is another matter: these are the explicit rules (ToS, rules of stakeholder behavior that designers come up with). This moderates behavior on the platform at the user and ecosystem level.
    • Underlying principle for governance: Give fast, open feedback when applying laws that define good behavior, but slow, opaque feedback when applying laws that punish bad behavior.
    • Generally, these platform laws should be transparent.
      • 💼 Stack Overflow has explicit list of rules for earning points, and the rights/privileges that these points award to users. Points gain you the right to ask/answer questions, upvote/downvote someone's content, see fewer ads. This list of rules and rewards creates a public good by encouraging users to be good actors and share good insights.
    • Sometimes these rules shouldn't be transparent, because it may facilitate bad behavior.
      • 💼 Dating sites: When stalkers understand what is considered "bad behavior" on a platform, they learn to game it. Platforms that delay the negative consequences of certain behaviors result in stalkers having a harder time identifying the trigger and working around the system.

🔑 Tool 2: Norms

  • Vibrant communities are a huge asset to a platform, and these are generally nurtured by skilled managers that develop norms, culture, and expectations that generate lasting sources of value.
  • 💼 iStockphoto
    • Originally a business that sold CD-ROM collections of images by direct mail. Wasn't going well, so they started giving away images online. Users discovered them, wanted to use the images and also upload their own.
    • In order to maintain quality, every image was scrutinized by an inspector. This was costly, couldn't scale.
    • They turned to crowd curation. People uploading content could earn their way into becoming inspectors, had specific topic spaces they inspected. CEO gave a lot of praise and feedback to these users. The norms established here led to a huge amount of high quality stock photos, which is a public good.
    • The established norms in the community included:
      • Feedback
      • High quality content
      • Open engagement
      • Natural role progression to greater levels of authority
  • Norms can be created through behavior design.
    • 🔖 Behavior design = a recurring sequence of trigger, action, reward, and investment.
      • Trigger: platform based signal, message, or alert (email, web link, news item, notification)
      • Action: The trigger prompts platform member to take an action.
      • Reward: The action produces a reward for the member. Bonus if it's a variable/unexpected value, which creates habit-forming behaviors.
      • Investment: the platform then asks the user for an investment (time, data, social capital, money), which deepens the participant's commitment and reinforces behavior patterns.
    • 💼 Behavior design example: Pinterest
      • Someone sees an interesting photo on Facebook (the trigger), clicks on it (the action, should be frictionless) and lands on Pinterest. Their reward is all the good content on Pinterest related to that photo that is curated to be interesting to them. Then Pinterest asks for an investment, like inviting friends, giving the user preferences, asking the user to learn new features. This sets up new triggers and the process starts again.
    • Behavior design can result in a public good or value, but it can also be manipulative and doesn't always benefit participants. This is why users should understand how governance mechanisms work. It's important to have users participate in shaping the systems that govern them.
      • Communities that successfully create and police public goods tend to have several regular patterns. This was an observation by Elinor Ostrom, the first woman to receive the Nobel Prize in Economics. These patterns are:
          1. Clearly defined boundaries between who has access to community benefits and who doesn't.
          1. People affected by how resources are used are given recognized channels they can use to influence the decision-making process
          1. Community monitors are accountable to the community.
          1. Various levels of increasing sanctions are applied to those who violate community rules.
          1. Members can use low-cost dispute resolution systems
          1. As community resources scale, governance should be in a series of nested tiers: simple issues controlled by small, local user groups, while complex/global problems managed by more formally organized groups.
        • 💼 eBay
          • Friction between buyers and sellers when trying to update the traditional auction format. Buyers liked fixed prices, sellers didn't. To resolve, used a number of Ostrom's ideas:
            • Focus groups and a "voices" program to reflect user perspectives and gauge the priority of their feelings.
            • Thoughtful communications about proposed changes
            • Tested programs on smaller groups, pulled back on bad changes
          • Ended up siding with the buyers, reasoning that sellers would remain loyal to the platform since that's where the buyers are. Today, "buy it now" fixed-price listings account for 70 percent of eBay's gross merchandise volume.

🔑 Tool 3: Architecture

  • In a platform business, architecture = programming code.
  • "Well-designed software systems are self-improving: they encourage and reward good behavior, thereby producing more of the same."
    • 💼 Online banking platforms: peer-to-peer lending
      • These businesses use algorithms to displace traditional labor-intensive loan officers. They use conventional (credit scores) and unconventional (Yelp ratings, online connections) data to calculate a borrower's likelihood of repaying. As this algorithm gets better, risk declines, which attracts more lenders. Low overhead costs = low rates = more borrowers.
  • Architecture can be used to prevent and correct market failures.
    • 💼 eBay
      • The middlemen who exploited the inexperienced sellers (previous eBay case study) were problematic, but they did provide market liquidity known as arbitrage. Without these arbitrageurs, nothing would happen, so they're providing a service while also highlighting a market inefficiency. To deal with this, eBay created automated systems like spelling assistance to provide confidence that sellers will receive an item's worth.
      • This is good governance that may disenfranchise the arbitrageurs, but increases the overall health of the system.
    • 💼 Bitcoin
      • In 2008, Satoshi Nakamoto published a paper on Cryptography, defined the Bitcoin digital currency and the blockchain protocol governing it.
      • Blockchain is revolutionary: it allows for decentralized, completely trustworthy interactions without any need for escrow payments or other guarantees.
      • 🔖 Blockchain: distributed public ledger that enables storage of data in a container (the block) affixed to other containers (the chain).
      • The resulting smart, autonomous contracts execute the rules automatically. An exchange of services and payment are ensured without either party controlling the system.
      • This has led to a new kind of platform: open architecture and a governance model but no central authority. Also addresses monopoly power - no one can remake these rules to favor one participant over another.

🔑 Tool 4: Markets

  • Markets can govern behavior using human motivations like fun, fame, and fortune. Social currency is often more valuable than money.
    • Ex. Providing a good interaction, quality content, and social network — a foundation for reputation.
  • 💼 iStockphoto
    • Provided virtual currency in form of photo download credits. Original uploader of photo receives the cost of a download at 1 credit. Credits could be purchased with money, and uploaders could cash out accumulated credits. Professional and nonprofessional photographers could both benefit from participating in the same market.
  • 💼 SAP (Enterprise Management Platform Company)
    • To motivate developers to answer one another's questions: when someone from a company answers a question, points are credited to the company's account. At a certain level of points, SAP will make a contribution to a charity of the company's choice. That has saved SAP $6-8 million in tech support costs, created new product ideas, and reduced question response times. Accounts for a $500,000 gain in annual productivity per enterprise software partner.
    • Social currency used to stimulate SAP's developer economy: When SAP introduced a new CRM, it offered double points for any answers, code, or white paper related to the CRM. In 2 months, this led to developers finding gaps in the software and developing new features.
  • Given that useful ideas are public goods, what's the best intellectual property policy for a platform business? When developers on a platform produce valuable inventions, does ownership lie with the developer or the platform?
    • This is complicated. Ownership to developers = incentive for creation, but giving ownership to the platform facilitates standardization and sharing.
    • SAP had 2 practices to tackle the problem of ownership.
        1. Published a 2-year advance roadmap detailing new products and services. Served as a metaphorical patent period for outside developers.
        1. Partnered with developers financially (sometimes via acquisition). Ensured fair compensation, reduced partner risk, and encouraged outside investment in the SAP platform.
  • Risk reduction is an old problem; platform owners tend to avoid responsibility for risks faced by platform participants.
  • 💼 Fair Credit Reporting Act 1970
    • Background: Credit card companies avoided insuring cardholders; argued that insurance would promote fraud, resulting in careless cardholders and banks becoming more reluctant to extend credit, which would hurt low-income customers.
    • The amendments required fraud insurance and imposed a 50$ limit on consumer liability for fraud on their credit cards.
    • Without having to worry about fraud, consumers ended up using their cards much more.
    • This was so successful that many banks now waive the 50$ charge if a lost or stolen card is reported within a day.
  • 💼 Others
    • Airbnb refused to compensate hosts for bad guest behavior. Uber refused to insure riders for bad driver behavior.
    • Both companies recognized that this hurt growth of their platforms, and have introduced insurance.
  • Good governance: rather than minimize their own risk, platforms should reduce risks for their participants to maximize value creation.

🔑Principles of Smart self-governance

  • When platforms adopt the rules that apply to partners and participants, results improve.
    1. Internal transparency: clear and consistent communication / vision across divisions inside a company.
    • Internal departments in companies tend to become siloed and develop unique perspectives, languages, systems, etc. This makes it difficult to solve complex problems that require collaboration between two departments, or engage effectively with users and partners.
    • Platform managers need to deliver clear, consistent vision across the entire platform.
    • 🔖 The Yegge Rant was issued by Jeff Bezos of Amazon as a framework for clear communication. It mandates the exclusive use of service interfaces, which are data communication tools specifically designed to be clear, understandable, and useful internally and externally.
      1. All teams will henceforth expose their data and functionality through service interfaces.
      1. Teams must communicate with each other through these interfaces.
      1. There will be no other forms of interprocess communication allowed.
      1. This is technology-agnostic.
      1. All service interfaces must be designed from the group up and be exposed to developers to the outside world.
      1. Those who don't do this are fired.
      1. Have a nice day!
  • This helped Amazon consolidate its various web service operations to store, search, and communicate data to one operation with universally comprehensive protocols.
  • This had implications for utility in external applications for customers.
  • It is hard to scale if companies cannot see across their divisions.
  • 💼Sony
    • Walkman dominated portable music since 1970s. A few years after initial introduction, the iPhone swept the field.
    • Sony never developed a vision to offer a platform despite introducing the PlayStation Portable, owning Time Warner movie and TV studios.
    1. Participation: platforms should give external stakeholders an equal voice in internal decisions. Otherwise, bias towards favoring the platform over partners can lead to platform abandonment.
    • 💼USB promoted by Intel
      • First standard to facilitate data and power transfer between peripherals (keyboards, memory devices, monitors, etc)
      • However, peripherals were not inside Intel's core microchip business. Problem was chicken-egg; who produces peripherals for a standard no one has adopted, who buys a computer that doesn't have peripherals made for yet? Outsiders reluctant to partner with Intel, because Intel would have monopoly over the standard.
      • Intel entrusted this problem with IAL (Intel Architecture Labs), which was not controlled by any internal product. Served as neutral negotiator between ecosystem and internal business. Earned trust of partners by advocating for policies that benefited ecosystem at the expense of Intel business.
      • IAL invited over 50 companies to help define the standard and license. Intel committed to not trample partners' markets, used reputation and contracts to limit its own future behavior.
    1. Fair and just governance can create wealth: People treated fairly are more likely to share ideas. Diversity of ideas creates more ideas.
    • Fair governance leads to participants in a market to allocate their resources more wisely and productively; i.e. lead to a collaborative environment where entities are more likely to participate in anticipation of receiving a fair share of created value.
  • Governance will be imperfect because partners will find new forms of private advantage. When platforms allow third parties to innovate, platforms will struggle to control that value.
  • When conflicts of control arise, governance should favor greatest sources of new value or trend in the new direction; otherwise, face stagnation.
  • Governance systems should be flexible to respond to both fast and slow moving parts.
You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.