Rules as Code and Digital Identity: Highlights from BenCon 2023

Colorado Digital Service
6 min readSep 19, 2023


picture of Jeff Maher from Bloomworks, Alex Orlov from the Colorado Digital Service, and Tristan Vanech from the Colorado Digital Service
From left to right: Jeff Maher from Bloomworks, Alex Orlov from the Colorado Digital Service, and Tristan Vanech from the Colorado Digital Service


We’re Alex Orlov and Tristan Vanech, and we’re members of the Colorado Digital Service. Formed in 2019, CDS is a cross-functional team of senior engineers, human-centered designers, product managers, and procurement and contracting specialists within the Colorado Governor’s Office of Information Technology. Both of us are currently supporting the Behavioral Health Administration, which is working on making behavioral health services more accessible, meaningful, and trusted.

In June 2023, we embarked on an adventure that led us from the picturesque Colorado Front Range to the bustling corridors of power in Washington D.C.. We had the privilege of attending BenCon, a first-of-its-kind conference that brought together civic technologists, government employees, tech vendors, and thought leaders from around the world to focus on discussing how we can work together better to deliver public benefits. With only a few months of our Colorado Digital Service tour under our belts, we had no idea what to expect when we landed in Georgetown for the two-day event. After a delayed flight and a few hours of shut-eye, we were excited to immerse ourselves fully into the latest insights about the digital delivery of public benefits.

Follow more stories from CDS by signing up for our email list.

Day 1 Recap

The first day kicked off with a fascinating presentation and discussion of “Rules as Code.” If you’ve never heard of “Rules as Code,” this refers to the very nascent idea that the U.S. Federal government could author computer-readable code that would then be used by state governments and service providers. These “rules” (the code) could be easily updated in multiple systems, all in one go. This type of intentional rules-as-code authorship would reduce the burden and ambiguity for governmental employees and institutions that are interpreting and administering benefits and eligibility systems. Today these rules are frequently changing and it is hard to quickly reflect those changes in the technologies deployed for services. By using this model, governments could more easily model and measure the impacts of policy changes. On the beneficiary side, people would be able to more reliably predict benefit amounts and could more easily cross-enroll in different benefit programs, saving time and reducing duplicative data entry.

The session on Rules as Code left us pondering how federal and state governments are inextricably linked when it comes to many benefits programs like the Supplemental Nutrition Assistance Program (SNAP) and Temporary Assistance for Needy Families (TANF). When it comes to digital delivery of benefits, how can these entities use technology to establish more efficient processes when it comes to legislation changes?

Our other favorite session of Day 1 was about the future of digital identity. If you’ve ever had to answer security questions to log into a bank account application, you’ve used a technology product that is trying to figure out if you are who you say you are. Although there are some promising paths to verifying digital identity for public benefits programs, identity validation remains a big point of friction for many people in America trying to access public benefits. There is an inherent tension between creating processes and technologies that focus on fraud protection and approaches that seek to improve the accessibility of benefits. An example we heard really drove this point home: A 30% pass rate for identity proofing might look great to someone trying to prevent fraud, but terrible to a legal services advocate whose clients are eligible for benefits but unable to pass through the identity proofing gates. This means that while fraud was prevented, it comes at the cost of receiving the benefits. The conference speaker, Elizabeth Bynum Sorrell, advocated for public benefits programs giving users a choice between multiple methods (biometrics, in-person verification, etc.) for verifying their identity.

Jen Pahlka, author of of the book Recording America: Why Government is Failing in the Digital Age and How We Can Do Better, stands in front of a slide that compares “Doing exactly what the statue says” and “Doing what it intends” to illustrate the difference between simply creating solutions based on the words of legislation versus the impact it intended
Jen Pahlka gives a keynote discussing the difference between simply creating solutions based on the words of legislation versus the impact it intended.

Day 2 Recap

The second day started with a panel discussion about equity in digital benefits technology. We learned about using text messages as outreach for beneficiaries, conducting design research on topics like child welfare systems, and ideas around using data ethically.

The morning ended with a keynote talk from Jennifer Pahlka, who discussed how the government has been “starved of” good design and technology. She shared an alarming anecdote about helping California reduce their unemployment insurance backlog during COVID — it took an ex-Google engineer WEEKS to merely measure how bad the backlog was. That’s how disorganized the system was!

To meet policy goals, Pahlka said that governments must focus on implementation and design by investing in people first and technology second. She also mentioned that traditional government tech teams often suffer from “procedure fetish” where they follow exactly what statute says rather than what it intends. Most of the time laws are written with good intentions, however how these intentions can be misinterpreted or misapplied.. Uncovering those good intentions can help us nail down the spirit of the law so we can interpret them correctly and serve the needs of the constituents they were written to serve.

Key Themes

Throughout the conference, many speakers reiterated the importance of designing with and for people — beneficiaries as well as government employees who need to navigate complex systems on a daily basis. Even in our digital age, many beneficiaries still rely on face time with government employees in order to access services. There was a strong emphasis on user-centered design and the ways in which technology can both help and hurt beneficiaries. We left the conference with a deeper understanding of the burden of enrolling and maintaining eligibility in benefits programs.

Our Reflections

Both of us were blown away by the quality and quantity of information during the two-day event! Plus, we loved making connections with others. We learned that the civic tech community is small and big at the same time. It’s small in that a lot of folks know each other; it’s significant in that there was a wide variety of roles and responsibilities. We met people working in federal government roles, people working in state government roles, vendors, people at think tanks, people who are lawyers representing people who can’t access benefits, and so many others.

Some of the most eye-opening insights came during informal chats with other conference attendees. Someone who worked in the SNAP space shared that beneficiaries could receive as little as $23/month. Delivering that benefit can’t possibly be efficient or worth anyone’s time, even the beneficiary’s! It got us thinking about the history of the social safety net and how it was designed to have people fall through the cracks. Although basic income was mentioned as one of six benefits the conference was centered on, there was little discussion of it. Our hypothesis is that there isn’t much digital delivery work in the UBI space yet.

What we’re still chewing on: How does the presence or absence of means testing (e.g. income-based assessments) reflect the intent of legislators, and how do digital tools help or hinder that intent? Could universal benefits that eliminate means testing correct for some of the ragged edges in our system? What could a new world of public benefits look like with fewer administrative burdens, so systems have less reliance on means testing and people get more direct help when they need it?

This article was written by Alex Orlov and Tristan Vanech.

Interested in following along with the work of the Colorado Digital Service? Submit an interest form to stay connected. You can also check out our work at our Colorado Digital Service website, GitHub, Linkedin, or Twitter.



Colorado Digital Service

A team of engineers, designers, product managers, and procurement specialists serving limited tours of service in government.