The Silent Failure of Digital Government

November 30, 2020

Close Up of a Government Building

The Origins of Dysfunction in Online Government Services and the Road to a Cure. 

A couple of months ago, our Founder & CEO Harish R. Rao, gave a keynote address at Drupal GovCon 2020, a virtual gathering for developers and managers of government websites. In that address, he shared a shocking statistic derived from the data we gathered from roughly 15 million government and municipal website users across the country. Nearly one out of three visitors to government websites fail to complete the purpose of their visit.

Not surprisingly, users’ failure rates went up even further as the Covid-19 pandemic worsened and the government at all levels scrambled to adapt more services to online platforms. How did things get this bad? How can they get better?

The answer to the first question has a lot to do with the ways government sites are assessing their own performance. For nearly two decades now, government sites have been measuring their performance based on metrics that may seem useful but in reality, could actually be undermining their users’ success. Specifically, the number of page views (how many visitors), bounce rates (a measure of how “sticky” a website is), and session duration (time spent on the site). All three of these metrics are default industry standards enshrined by Google Analytics, the industry standard. Although they now seem indisputable, they were born out of specific circumstances that bear virtually no relevance to the purpose of government sites.

Google Analytics itself evolved out of Urchin, a late-nineties and early-aughts web file analysis software. As Urchin/Google Analytics evolved into the standard-bearer for tracking online browsing data, they were catering to very specific use cases: online publishers of periodicals and the budding realm of e-commerce. The metrics Urchin/Google Analytics zeroed in on perfecting were geared to these contexts. Page views would help publishers know how much they could charge for ads, not unlike the formulas they’d used in their print publications. Bounce rate and time on page, aided by innovations like tracking pixels, would help e-commerce sites convert visitors into customers by learning how to keep them browsing for longer periods of time and how to nudge them to complete their transactions.

Urchin and Google Analytics were both, for good reason, wildly successful. And Google Analytics continues to have more than half of the market for web analytics. Urchin was so successful because it provided measurement and analysis for the then burgeoning web, and Google Analytics continues to innovate and provide a virtually free tool for website creators today.

Neither of these use cases has anything to do with good government. At their core, government websites have three key use cases: finding information, accessing services, and making payments. Within this context, the Google Analytics metrics become not only meaningless but detrimental. For example:

  • A higher time on page might signify confusion, stagnation, or technical problems rather than an engaged customer interaction. It’s the digital equivalent of being stuck in a line at the DMV.

  • Bounce rate doesn’t matter if it means the user quickly found the information they were looking for.

  • And while more page views might mean more people accessing online information and services, it doesn’t tell us anything about what experience those users are having once they arrive.

If we want to make government websites better and online access to essential services more streamlined and efficient, we need to stop using the wrong data. Instead of stale metrics designed for an entirely different context, we should adopt new government-geared metrics that tell us more about where individuals are succeeding, where they’re getting frustrated, and how they feel about their overall interactions. At I.F., we call these metrics user success and user satisfaction, and they’re at the heart of our Voice of Citizen® analytics platform. We track them for millions of people using government sites across the country so we can learn what’s working, what’s not, and where we can effectively intervene.

But you do not need a proprietary analytics platform to measure user success. A simple poll, “were you able to complete the purpose of your visit today?” is a great start.

The first step to addressing this silent unacceptable failure across government sites is to start asking the right questions and listening to the answers. You don’t even need a full analytics platform like Voice of Citizen® to start looking at more meaningful metrics. A simple two-question survey can get you started.

Question 1: Were you able to complete the purpose of your visit today?

Question 2: If not, why? Then turn the responses you gather into a plan of action to address the failure points. Keep measuring the impact of your efforts. Keep fine-tuning your solutions.

This user success data should be more than just a shocking statistic — it should be a wake up call, a rallying cry. We’ve come to expect that government services won’t be great. We’ve normalized that a good percentage of users will struggle to achieve their goals or eventually just give up. It doesn’t have to be this way. We can change it together.

The first step is raising our expectations and standards. Government sites should be better than just “good enough for government work.” We should expect — even demand — something truly useful, accessible, and relevant to our real needs. Only then can we cure digital failure and reap the benefits of a government that’s more than “good enough.” Let’s get to work.