Fixing the FAFSA With Data, Testing and Transparency
Singer: After a difficult rollout, the U.S. Education Department has produced a better tool to help students secure financial aid for college.

Get stories like this delivered straight to your inbox. Sign up for The 74 Newsletter
In November, the 2025-26 Free Application for Federal Student Aid (FAFSA) launched to all students and families. While far from perfect, the FAFSA will enable over 16 million students to secure more than $100 billion in federal aid, helping make higher education accessible for millions of lower-income and first-generation Americans.
The challenges with last year’s launch have been well-documented. In fact, I took a six-month leave from my post as president of the College Board to join the U.S. Department of Education in fixing the FAFSA. After months of hard work, we are confident this year will be much better. We arrived at this point by adhering to a few best practices for technology development, whether for the government or the private sector.
The development and launch of a new FAFSA for the 2024-25 admissions cycle faced unique challenges. In today’s world, policy complexity produces software complexity. In turn, software complexity produces increased expenses, delays, and errors.
In December 2020, Congress passed the FAFSA Simplification Act – well-intended, bipartisan legislation that aimed to shorten the form, increase accuracy by importing tax data from the Internal Revenue Service, and, ultimately, expand eligibility for Pell Grants.
This Act required brand-new software composed of multiple components that are themselves integrated to a dozen other systems, many housed in other agencies. While the Department of Education is full of devoted public servants with tremendous knowledge about financial aid and a passion for helping people go to college, they had limited experience building modern, complex software applications. In addition, due to the nature of government contracting, the Department ended up working with four separate vendors to build the new FAFSA.
Given all of these challenges, the problems encountered during the 2024-25 launch should not have been a surprise. But that is little comfort to students and families who had a difficult experience last year.
So why is this year better? In preparing for the 2025-26 launch, our work was guided by five key principles, all of which can be applied to any government endeavor:
- Focus on what is most important. A common mistake in software is trying to do too much. Failing to make tough choices leads to missed deadlines, poor code quality and user frustration. We focused on two things: 1) fixing any bugs that prevented students from submitting the form; and 2) delivering a stable application on a defined timeline for students, families, and institutions of higher education. This required us to make tough choices, such as deferring some new features, which allowed us to launch well ahead of last year’s schedule.
- Use data to identify critical user issues. Data is a powerful tool to zero in on challenges that affect large numbers of users. For example, a question in the 2024-25 FAFSA about a “direct unsubsidized loan” was originally worded in a way that unintentionally caused too many students – over 5% – to forgo Pell Grants and subsidized federal loans. After seeing the data and working with users, that question was redesigned, and the number of students misinterpreting it declined precipitously.
- Invest heavily in testing. Comprehensive testing is the most important aspect of any complex software project. We developed multiple new testing tools and methodologies, marrying data science and automated testing, that enabled us to verify the accuracy of the data we send to colleges, universities, and state agencies. We collaborated with the IRS to double-check the tax data that we receive. Most importantly, we conducted seven weeks of beta testing with more than 70 organizations – college access nonprofits, high schools and school districts, colleges and universities – before expanding testing to all interested students earlier this week. Over 67,000 students submitted real FAFSA forms during this testing period, and we traveled to many universities to sit with financial aid professionals as they independently verified the data we sent them.
- Embrace transparency. Because of the uncertainties and delays surrounding the 2024-25 FAFSA, key external stakeholders felt unprepared to support students. A chorus of advisors from the wider community told us that we needed to be more open about the work underway this year, both strides and setbacks, in order to build back this trust. We invested time and effort in building more channels and frequency for communication, including a new website at fafsa.gov/beta where we regularly updated statistics, provided updates on bug fixes, and shared stories from the field.
- Harness the power of the broader community. The success of the FAFSA depends not just on the government but also on a large ecosystem of organizations that play a vital role in supporting students and families. In part because of the breakdown of trust, the community was often out of sync and at times at-odds. By intentionally engaging these stakeholders, we were able to strengthen the partnerships needed to deliver a successful launch for families, including a critical testing period.
Ahead of the launch in November, we knew that the large majority of students can complete the FAFSA quickly: Over 90 percent of this year’s applicants reported that they completed it in a “reasonable amount of time.” Ultimately, for many people, the promise of FAFSA simplification – a simpler form that provides more aid – has been or will be realized.
There is more work to be done in the years ahead. There will be families who may struggle with the FAFSA; there are usability issues we have not had time to fix yet. For example, the ability for a student to invite their parent to the form needs to be simplified. The team is aware of these problems and will now turn its attention to making further improvements that will benefit everyone, especially those from underrepresented communities.
Building good software is hard; building good software under the unique constraints of the federal government is harder. But delivering a simpler FAFSA that serves students, families and institutions is a strong first step – and continuing to apply these lessons will build on that success.
Get stories like these delivered straight to your inbox. Sign up for The 74 Newsletter