What you will learn from reading The Platform Revolution:
– The characteristics of a platform business.
– What networks effects are and how they work.
– A list of proven acquisition strategies for bring users and producers onto a platform.
The Platform Revolution Book Summary
The Platform Revolution Book Summary is the complete guide to understanding platform companies. It dives into network effects, promotional tactics and all problems the platform businesses must overcome.
The Book Summary also covers monetisation strategies, launch strategies and platform governance. If you’re interested in companies such as Uber, Air BnB and even Amazon have built successfully, then this is the book for you.
What are Platforms?
A platform is a business based on enabling value creating interactions between external producers and consumers.
Proxies needed for algorithims to connect.
Platforms change value chains and create eo-systems:
Traditional business are pipelines step by step arrangement for creating and transferring value. Linear value chain.
Platforms beat pipelines because platforms scale more efficiently by eliminating gatekeepers. Gatekeepers time consuming and inefficient e.g. editors for publishing.
Platforms unlock new sources of value creation and supply. (Removes industry limitations, such as supply of value creators)
Platforms foster new supply to the market. Community used to be source of demand but with a platform can also be source of supply. Your community becomes an asset.
How to Manage Platform Behaviour:
Using default insurance contracts and reputation systems incentivises good behaviour and fosters credibility and trust.
Platforms used data-based tools to create community feedback loops.
You need to leverage the community to grow and police the content. Show what is allowed and what isn’t.
Platfroms invert the firm – shifts its focus from internal activities to external.
Networks effects can be both positive and negative.
Disruptive technological change renders old industry economics obsolete.
Demand economies of scale are fundamental source of positive network effects.
Convex growth and collapse can occur with network effects. Therefore, acquisition and retention key.
Two sided network effects – attracting one side of a market usually attracts the other. However, asymmetrical value in proportions.
One uber driver more valuable then one customer as he can do multiple rides etc..
Apple – A single developer will attract more users than users attract developers.
How to Speed up Network Effects:
Is fool-proof plan for temporary enhanced market share. What percentage of free users become buyers is small. Requires high volume.
Most companies compete with price effects and branding effects, these aren’t as powerful as network effects. Remember – Algorithms scale better than employees.
Frictionless entry is the ability of users to quickly and easily join a platform and begin participating in value creation that the platform facilitates which is a key factor for growth.
Scalable business models require frictionless entry and side switching to successfully lubricate network effects.
Curation of Feed or Good Matching:
Effective curation can be used to prevent negative network effects. Filters, limits and controls access of users on platform.
Strong curated algorithms need to produce good matches, this requires more data for better accuracy.
Focus on happy matches through curation = more value generated.
Remember — Poor curation leads to greater ‘noise’
Growth comes not from horizontal or vertical integration but functional integration and network orchestration.
Two types of network effects:
- Same-side effects
- Cross-side effects
Interactions on a platfrom resemble any economic or social exchange thus, they exchange 3 things.
- Goods or services
Attention, fame, influence, reputation and other intangible forms of value play the role of currency. (Social Currency)
Every platform should start with the design of the core interaction.
The core interaction has 3 components:
- The participants.
- The value unit.
- The filter.
Decisions need to be made on who can create value units, how can they be created and integrated into a platform and what differentiates high quality from low quality.
The roles need to be explicitly described and understood.
The same user may play a different role in differing interactions. A well designed platform makes it easy for users to move from role to role.
The value unit:
Air BnB and Ebay value is listing information served to consumers based on search or past interests.
Provides information to help decide whether to proceed to some further exchange.
Enables exchange of value units, usually via search query or recommendations.
The kind of interactions involved will grow as people work out what’s possible. This will likely create more problems to solve!
Platforms are ‘information factories’, they create the ‘factory floor’ and can foster a culture of quality control (incentives).
Key Concepts in Platforms:
Pull users – Consumers and producers to the platform
Facilitate interaction – Providing tools and rules to make easy connections and valuable exchanges
Match users – Effectively using information to connect in a mutually rewarding way.
Create Feedback loops – Recommend things of interest to keep users coming back (curation).
Techniques for these concepts:
Pulls can leverage outside networks (such as signing in with FB or sharing posts across different social media sites)
Facilitating can involve removing barriers or increasing them. I.e. making a more credible service by increasing regulations e.g platform for babysitting
Companies need to develop an explicit data acquisition strategy.
LinkedIn – Gamifiy experience by using a progress bar
Leverage by signing in through other accounts. Gives more data.
Several ways core interaction can be added to:
- Change value unit exchanged between existing users
- New category of users as either producers or consumers (advertisers added)
- Allowing users to exchange new value units
- Curating a select group of current users to create a new category (thought leaders)
End to end principle – application specific features should reside in the outer layer of a platform, rather than deep in the roots.
Stable core sitting underneath an evolving layer that facilitates variety.
Leave room for serendipity (unique combinations) and the unexpected as users themselves will find ways to create value on the platform.
‘The internet no longer acts as a distribution channel but is also a creation infrastructure and co-ordination mechanism.’
Platforms flip from internal resources to external resources. Benefiting from superior economies of production and distribution.
The Democratisation of tools of design has led to disruption.
A great question to ask — What new empowering technology can be leveraged?
Curation key – algorithms can collate community judgements.
Structural impacts of platform disruption:
De-linking assets from value
De-linking ownership of the physical asset from the value it creates.
Example: MRI machines cost millions to purchase yet get used 50% of what it could. Therefore timesharing to those that can’t afford to pay for one would add incremental revenue to the business plus get proper usage out of the machine.
Re-intermediation – bringing back middleman in the form of algorithms and social feedback.
Launching a Platform:
Platform Launch Strategies:
- Follow the rabbit – Change from a successful pipeline to platform
- Piggyback – Connect with an existing user base from a different platform and stage creation of value units in order to recruit (paypal with eBay)
- Seeding – Value creation by acting as the first producer. Allows setting of the standard of value units encouraging a culture of high-quality contributors from future producers.
- Marquee – Provide incentives to attract members of key user set onto your platform
- Single-side – Create a business around products or services that benefit one side, later convert the business into a platform and attract the other side.
- Producer evangelism – Design your platform to attract producers, who can introduce their customers (followers) to become users of the platform
- Big bang adoption – Push strategies to attract high volume of interest and attention
- Micro-market – Target a tiny market that comprises of members already engaging in interactions (FB’s start with just campuses)
Convert users into marketers to get explosive growth
In general, users spread self-created value units to get social feedback.
Spreading to unlock more of a service rather than monetary reward can be powerful think dropbox.
4 broad categories of value:
For consumers: Access to value created on the platform
For producers: Access to community/market
Both: Access to tools and services that facilitate interaction
Both: Access to curation mechanisms
Charging has an impact on network effects and creates friction.
Networks effects measured by the number of visitors doesn’t necessarily reflect the monetary value of a platform.
Take steps to promote desirable interactions.
Monetising interactions means tools must be created that benefit both parties by removing friction, mitigating risk and facilitating valuable interactions.
Problems with services:
- Meet in person
- Quality is less variable
- The consumer can monitor service provided directly
Enhance access: means being able to place self at top of a list can, however, increase noise level creating negative network effects.
Rarely charge all users.
Charge one whilst subsidising other
Charge most users full price whilst subsidising stars
Charge some pull price and subsidise those that are price sensitive
“You never take money first”
Keep as many monetisation options as possible open for as long as possible.
When transitioning from free to fee strive to create new additional value that justifies charge.
Platforms have two parts, its structure and operation. The operation (managers) directly touch the users. The sponsor (structure) retains legal control over the technology.
In the quest for proprietary control of a market you’re better off being as fast as possible before the next big thing supersedes the technology you seek to dominate.
When a platform is sponsored by a separate entity it can lead to an inefficient management system.
Core developers (usually employed by the company, core infrastructure)
Extension developers (app developers for the system, usually through API)
Data aggregators (Accumulate and sell data on to advertisers)
Extension developers can add extra value to the interactions such as airbnb photography service. However, excessive openness leads to too many extension developers which in turn because of competition reduces their margins making developers less attracted to the platform.
API – application programming interface:
Basically routines, protocols and tools for building software applications that make it easy for outside programmers to write code that will connect seamlessly with platform infrastructure.
Openness needs to be limited however community can provide screening and feedback which is helpful for effective curation. (Upheld by community standards and social pressure)
Platforms operating in similar areas can differentiate by adopting different levels and kinds of openness. Creates distinct ecosystem cultures.
API’s can be used effectively to monetary extension developers. Can restrict competition.
Openness needs to be questioned throughout the life of the platform.
3 fundamental rules of governance:
- Always create value for consumers you serve
- Don’t user power to change rules in your favour
- Don’t take more than you fair share
4 main causes of market failures:
- Information asymmetry (counterfeits)
- Externalities (spillover of benefits to those that aren’t involved)
- Monopoly power (power sellers on platform)
- Risk (tools to mitigate risk / insurance etc..)
Positive externalities include using data aggregated from other users to give more accurate recommendations.
Tools for governance include Laws, norms, architecture and markets.
Economist Alvin Roth designed a model of governance that has 4 broad levers:
1. Safety of market via transparency, quality or insurance leading to good interactions.
2. Thickness – enables participants from different sides to find each other more easily.
3. Minimise congestion when too many people participate or low quality drives out high.
4. Minimise repugnant activity such as forbidding porn.
Platform laws should be transparent
An exception to transparency is if it facilitates bad behaviour such as loopholes. Give fast open feedback on laws that define good behaviour and give slow opaque feedback to laws that punish bad behaviour.
Crowd curation – role progression to greater levels of authority (responsibility)
Behaviour design is the recurring sequence displayed by the platform users as discussed by Nir Eyal: Trigger, action, reward, investment (ask).
Variable reward mechanisms are habit-forming (think gambling).
Well, designed software systems are self-improving.
Mechanisms should be designed into it to offer various incentives not just monetary… fun, fame and fortune. Subjective forms of value (social currency)
Can give metaphorical patents to extension developers, also have a policy of partnering with developers financially or buying out at a fair price.
Good governance means looking after the health of one’s ecosystems partners.
Principles of smart self-governance:
- Internal transparency between departments
- Participation – give external partners and stakeholders a voice in internal decision processes.
- Just and fair governance can create wealth
Governance must be self-healing and promote evolution. It is not static.
Platform metrics need to measure the role of interaction success and factors that contribute to it.
Creation of value most important to encourage continual growth of positive network effects.
Most tools from general management aren’t designed to flourish in the harsh soil of extreme uncertainty which start-ups thrive.
3 main metrics for start up:
- Matching quality
A reasonable way to measure liquidity is by tracking the percentage of listings that lead to interactions within a given time period.
Indicates engagement (when a user has recognised or used a value unit)
*Meaningful metrics are comparative ones, helps by making distinctions. Such as the rate of growth of active users or ratio of active users. (actionable not vanity metrics)
Matching quality lies in curation – sales conversion rate?
Trust is developed through excellent curation which gets better with more data.
Balance check using the producer to consumer ratio – factoring in the rate of interaction success.
On the producer side:
- Frequency of producer participation
- Listings created
- Outcomes achieved
- Monitor interaction failure
- Producer fraud
LTV can now be used.
- Frequency of consumption
- Rate of conversion to sale
LTV also can now be calculated
Run experiments to identify the critical determinants of LTV e.g. churn rate
Monitored metrics can help develop smart strategies / extension ideas
Switching rate also an important metric.
Distance between consumers and producers which is the frequency of direct interaction, the six and reach of networks can be roughly worked out by the size of the advertising budget. Thus, industries with high advertising budget may benefit from platforms?
- Drive innovation
- Have a high signal to noise ratio
- Facilitate resource allocation
Identify necessary adaptions by studying extensions provided by developers.
Don’t be overmeasure and under-prioritised (80/20)
Particular effective barrier to entry is control of an indispensable and inimitable resource.
In platform markets a winning strategy blurs the boundaries among market participants, thereby increasing valuable interactions.
The resource pool is capable of growing much faster than the platform company.
Platforms seek exclusive access to essential assets. They should discourage multi-homing.
Platform envelopment – one platform effectively absorbs functions of an adjacent platform.
- Supply economies of scale
- Strong network effects
- High multi-homing/switching costs
- Lack of niche specialisation
Platforms – The Businesses of Tomorrow
Types of business susceptible to platforms:
- Information intensive industries
- Industries with non-scalable gatekeepers
- Highly fragmented industries (aggregation)
- Industries with extreme info asymmetries
- High regulatory control
- High failure costs
- Resource-intensive industries