Engineering smaller burdens
One area that holds enormous promise for helping realize this goal is improving the performance of government websites and digital services. Engineering decisions that go into building the front and back end of digital services can have significant implications for people’s experience and the time it takes them to use a service.
The performance of a website or digital service has direct implications for the size of the burden placed on users. Reducing the size of this burden requires agencies to think carefully and to engineer services to make potential burdens as small as possible.
A policy focus on reducing burdens
In late 2021, the Biden Administration issued an Executive Order directing agencies to take steps to improve customer experience on federal agency websites. More recently, the Office of Information and Regulatory Affairs launched an initiative to reduce both the amount of time and the level of effort required to interact with federal agency websites and online services. Taken together, these initiatives, and several others the Administration is pursuing, seek to reduce what is referred to as the “administrative burden” placed on people seeking to access online government services.
Administrative burden can take many forms. Guidance issued by the Office of Management and Budget last year directs agencies to consider not just the length of time it takes people to use online services but also to calculate the learning costs, compliance costs, and psychological costs different groups of people face when using government services. This guidance recognizes that many different factors impact people’s experience using online services and that any one of them may present a barrier that makes it harder for people to access benefits that they may need and to which they are entitled.
An often overlooked factor that contributes significantly to the experience of people using government digital services is the technical infrastructure underpinning those services. This infrastructure can be grouped into what software developers refer to as the “front end,” which is the web content delivered to a user’s browser, and the “back end,” which are the technical components that receive data (submitted from users through things like web forms), process and store that data, and interact with other systems to provide a government service.
Fixing the front end
A key facet of reducing administrative burdens on those who use digital services is to make changes to the front end – the part of the online service that a user sees in their web or mobile browser. There are a number of steps that agencies can take to reduce burdens in this part of a service:
- Make the web content more accessible
- Use plain language to present information and describe requirements
- Provide content in multiple languages
- Ensure websites display properly on mobile devices
But an often underappreciated aspect of reducing burdensome qualities on the front end of digital services is to simply reduce the size of the front end – make it smaller.
As web browsers and mobile devices become more powerful, it becomes possible to engineer an increasing amount of complexity into the front end of digital services. This complexity can support new features and enhanced functionality. But this increased complexity can come with a price – one that is paid more steeply by people using older computers and less powerful mobile devices.
To deliver more complex logic as part of the front end, we need to increase the size of the web content that gets sent to a user’s browser or mobile device. While this larger payload holds the promise of enhanced features for users with newer, more powerful devices, for others it means more time waiting. A larger bundle of web content takes longer to deliver to people with slower connections and older devices. Because the distribution of access to more powerful devices and higher-speed internet access is uneven, engineering more complexity into the front end means we will make some people wait longer than others. This is not only burdensome, it’s also unfair.
While there is nothing inherently wrong with leveraging the increasing sophistication of web browsers and mobile devices, the use of this approach should be part of a strategy that focuses on overall performance. Reducing the size of the payload a website or digital service sends to a user reduces the time that user needs to wait for a page to load. Enhancing the performance of websites and digital services reduces administrative burdens by reducing the amount of time it takes to access a service, and makes it easier for more people to use it.
Bolstering the back end
The technical infrastructure that makes up the back end of digital services can also contribute to the time it takes for a user to access a website or digital service.
These back-end components accept data submitted by a website user, process and store that data, and work with other systems to support an online interaction with a government agency. Depending on the number of different systems a digital service needs to interact with and the age or complexity of those systems, these back-end components can be quite complex. Any one of these other services that a digital service needs to interact with can create a bottleneck that can slow down a digital service, adding to the time a user will wait for their transaction to complete.
Increasingly, government agencies are moving the back-end components of digital services to the cloud. A key feature of cloud-based platforms is the ability to scale computing capacity up or down depending on the number of people using a system at any given time. However, not all agencies have fully migrated their services to the cloud. And for those that have, not all are effectively leveraging the ability to scale out capacity afforded by cloud providers.
The impact of back-end engineering decisions can have a significant impact on the burden placed on users of digital services. At no time was this point made more obvious than when states’ unemployment insurance systems were used as a way to deliver benefits to people whom the COVID-19 pandemic impacted. Faced with a dramatic increase in demand, these systems buckled under the pressure and millions of people were impacted either by substantially increased wait times to acquire benefits or by giving up altogether and forgoing needed benefits.
One study of the performance of state unemployment benefit systems during the pandemic by the Economic Policy Institute found that an estimated 10 million people may have been discouraged from applying for benefits during the pandemic simply because the process was so difficult and lengthy.
A unique feature of engineering decisions affecting back-end systems that can add to administrative burdens is that they may not always create problems for users. A back-end system that is designed to accommodate a small number of users may work sufficiently well in the absence of heavy demand. But as circumstances change and demand for information or services increases, these systems will slow down (or completely fall over) from increased traffic, adding to the burden of everyone who uses a digital service.
Just as the waiting area in a hospital emergency room might seem appropriately sized during normal times, if a natural disaster or other unforeseen event occurs, it suddenly becomes far too small. These design decisions are “situationally burdensome” – they may impose only a small or acceptable amount of burden during normal times, but when circumstances change suddenly, they can impose a huge burden on those who desperately need them. And while adjusting the size of a hospital emergency room after it’s been built is not practical, adjusting the computing capacity supporting a digital service is easy with modern cloud computing platforms.
But only if agencies plan ahead.
Keeping burdens small
Government agencies must do better than simply engineering digital services for “normal” times. Resiliency needs to be built into digital services to ensure that even when circumstances change, everyone who needs to access a service from a government agency can do so quickly and easily.
The current focus of the Administration on reducing burdens for people who need to access digital services is an important one. The engineering decisions that are made about the front end and back end of digital services play an important part in the burdens that those who need services may eventually face.
Related posts
- The Ad Hoc Incident Response Playbook: Helping you prepare for service emergencies
- How AI and LLMs will change government websites
- What AI and Large Language Models can bring to government
- Announcing the Platform Smells Playbook: Diagnose and fix common problems with your technology platform
- Introducing Ad Hoc’s Groundwork Delivery Platform
- Using open-source LLMs to optimize government data