This post is written by Lauren Lombardo, Master in Public Policy Student at Harvard University’s John F. Kennedy School of Government

  • **The problem: **There is no agreed upon way to define how a platform is architecturally designed or operationally implemented. This lack of common understanding makes it difficult to understand the benefits or costs of platforms.
  • Why it matters: The way we define these categories, the way we make choices within these categories, and the ecosystem (government structures) that we operate in all change the economic benefit and costs associated with platforms.
  • The solution: We need to use an architecturally-based definition of platforms and understand the variety of operational approaches and government structures that can influence the way a platform is built and implemented.

If you work in GovTech, the word “platform” has undoubtedly been the subject of many conversations. This is because platforms are frequently seen to improve government service delivery while reducing financial costs. They do this by capitalising on several economic properties, including economies of scale and complements.

But the more we talk about platforms, the less we seem to know how to define them — making it nearly impossible to build or use them responsibly. Realising the economic benefits of platforms first requires understanding what is and is not included.

**What is a platform? **

A traditional way to think about platforms is as a way to abstract technical infrastructure.

Abstraction reduces the amount of infrastructure directly managed by a government department and is therefore one reason platforms simplify service delivery. However, while a platform’s level of abstraction is important, it is not the only value add. And while platforms are abstraction tools, not all abstraction tools are platforms.

Therefore abstraction cannot be the defining factor. Instead, a platform should be defined by its architecture. An extensive review of platform literature, primarily from the fields of industrial economics and engineering design, coupled with dozens of practitioner interviews, revealed a three-part architectural framework that can be used for this definition.

A **platform ecosystem is a system of low and high-variety components with specified stable interfaces through which one accesses those components. **

The low-variety component** or the “core component” refers to highly reusable elements of the ecosystem that remain relatively stable and provide the main functionality. The interfaces **“govern the interactions of components,” are fully stable over time, and provide a way to control and access the entire platform ecosystem. And high variety components — or “complements” — are components or services that are peripheral to the core functionality and are encouraged to change over time.

Apple’s iPhone provides an easy-to-understand example of this three-part platform ecosystem.

  • Low-Volatility Component (Core): iOS, the system that provides the iPhone’s functionality.
  • Stable Interface: The App Store, a place for the market (iPhone users, app makers) to access iPhone components.
  • High-Volatility Component (Complements): Applications (e.g., Uber).

The changing economics of platforms

If a solution adheres to this three-part definition, it is a platform ecosystem. And platform ecosystems activate many economic properties, including economies of scale, economies of scope, combinatorial innovation, the economics of complements, and network effects.

However, the extent to which these economic properties are activated depends on the platform’s architecture, the operational approach, and existing government structures.

For example, for a solution to be a platform, it must match the architectural definition, but the way it meets the three criteria may vary. One platform ecosystem may have relatively open interfaces (e.g., internet) while another may have relatively closed ones (e.g., iOS). These variations differentiate the platform offering, change how a government can use it, and ultimately alter how strong certain economic properties are (e.g., in this case, the economics of complements).

A government’s operational approach can also influence these economic properties. There are three primary ways a government may operationalise a platform.

  • Use a platform: A department decides which platform to use without considering government-wide priorities. An individual department makes decisions and builds or buys platforms to meet its needs.
  • Share a platform: Departments frequently use a platform that is already used in other parts of the government, typically for core functions, such as hiring, payroll, authentication, or payments. Decisions about which platform to use are influenced by government-wide priorities and the number of other departments using the tool.
  • Be a platform: Most government functions exist within the same platform ecosystem. The low-volatility functions listed above (hiring, payroll, etc.) operate as the ecosystem's core components, which standardise them across all departments. Other high-volatility functions (department-specific operations) are provided by complements that sit on top of this platform.

If a government’s approach is to “be a platform ecosystem,” the two-sided or multi-sided market this creates will be subjected to network effects. Network effects are not as potent when following “share a platform ecosystem” or “use a platform ecosystem.” Similarly, economies of scope are stronger as more government services migrate to a single platform ecosystem, making “be a platform ecosystem” or “share a platform ecosystem” better transmitters of those economic benefits.

Regardless of the operational approach, other strategic decisions must be made. Will the platform be built in-house or provided by a commercial vendor? If built in-house, should it be cost-recoverable, subsidised, or free? Will the platform be mandatory for particular departments or optional? These decisions will further impact the economics of platform ecosystems.

Government structures

While architectural and operational decisions influence platform economics, government structures influence how those decisions are made. Government structures include procurement, budgeting, security, accountability and ownership, regulations, and political support. These structures create the actual and perceived support, limitations, and restrictions placed on platform ecosystems.

A government with strong accountability and ownership structures might find it challenging to implement a centralised “be a platform” approach. A government with strong security and regulations structures might find building a platform with open interfaces and loose complement rules too difficult.

Because these structures place guardrails on how a platform can be built and implemented, they alter how politically expedient, financially sound, or operationally feasible a platform ecosystem is.

Before implementing government platforms, decision-makers need to know how a platform is defined and what economic properties are at play. Further, decision-makers need to understand how their platform’s architectural design and operational approach, which are influenced by their own government structures, will impact those economic properties and ultimately alter economic benefits and costs.

Key takeaways

In short, platform ecosystems have nuanced architectural and operational specifications which correlate with real financial, functional, and political challenges that governments must acknowledge and address.

Platforms aren’t always the right solution. There are many government structures in place that can make platforms the wrong solution for some technical or operational problems. Even when a platform is a correct solution, these government structures can prevent the total capitalisation of benefits or, in some instances, create harm.

As a government looks to use platforms, it must understand how they may be built, used, and operated within its organisation and seek to understand what this means for economic and constituent impact.— Lauren Lombardo

This article has been adapted from the report Effective Implementation of Government Platforms: How architectural decisions, operational approaches, and government structures alter the economic benefits and costs of government platforms published by the Harvard Mossavar-Rahmani Center for Business and Government and the Harvard Ash Center for Democratic Governance and Innovation.

(Picture credit: Unsplash)