Live tools

Workforce Pell feels like a lot of hurrying up to wait for some and a mountain of data headaches for others, it seems. In conversations with workforce organizations, training providers, state education agencies, journalists, etc, we're seeing a few things:

1. Confusion

2. Optimism

3. Pessimism

4. More writing than tooling.

Some clear pain points have bubbled up, and this tool - lightweight as it is - is hoping to address some of them:

- "I am struggling to apply 1,000 pages of federal policy and murky state level data to our training program."

- "I don't know who to go to for what."

- "I don't understand what's happening at the state level."

- "We can't tell which programs are eligible for which reasons."

Now, did I spend too much time obsessing over the little piece of tape graphic that holds up the little state briefing cards for each state? Yes. But we find the joy where we can, ya know. At its core, the tool does these things: Check a program's eligibility for Workforce Pell and Provide a Briefing on the state of Workforce Pell for one State. I chose Louisiana because I live here and then New York, Florida, and California because they cover about 20% of workforce training providers (not programs) and have variance in how they're approaching the new federal rules.

Also, building out this tool got the most votes on labs, and it meets our 3 qualifications: someone asks for it, it could help workers, and it can be done in under 30 days. I did the building over 15 days. But the subject matter and user experience expertise side behind the build I'm not quite sure how to quantify. The majority of the work here was getting better and better at problem naming. So, I'll be really curious if it does actually solve any problems for you.

If it passes the desirability test, some ways to improve it are:

- The careers data. The wages data.

- The number of states available.

- Interoperability with existing systems.

Please feel free to chat with me about it - you can find me at leah@wherewego.org. Cheers.

I am not a mom, and I want to say that on the record. But over the past few years I've watched people in my family and community try to navigate pregnancy, postpartum, and caregiving alongside full-time jobs. Even in good situations, the logistics were exhausting — and the gap between what employers say and what work actually feels like for caregivers is wider than most people realize.

This tool tries to close a small piece of that gap. A scorecard. Sixteen questions. Deterministic scoring you can see through. It helps working moms, and anyone with caregiving responsibilities, figure out whether an employer is genuinely supportive or just claims to be — and surfaces specific things to ask for.

The research came first, synthesizing CDC/NIOSH guidance, federal protections like the PWFA and PUMP Act, and maternal retention studies. That synthesis is what shaped the categories, weights, and questions. The research is the product. From there it went through Claude Code and engineering review, and it'll get better once real working moms tell us what's missing. You can read more about the process here.

Software encodes assumptions. This is a small attempt to point that machinery the other way. Happy Mother's Day 💐!

If you've used a state's ETPL, you have a good understanding of what the gaps are. Trainingproviderresults.gov is a good wrapper on the ETPLs, but neither state ETPLs or TPR have effectively moved the needle on workers enrolling with training providers.

If you've used one of the many training navigators that are effectively a wrapper on CareerOneStop, you may have noticed some of their limitations. They rely on the same underlying search experience, with light customization to filters or branding. That approach is quick to stand up, but it inherits the same limitations. The relationship between the data and the worker or the data and their place is shallow, and the experience is not ready to adapt as much to a state or partner’s specific context as this is.

This tool is built for good data + good design + flexible customization for a state or sector's needs.

Scalable data

The system is designed to stand up quickly for a state while maintaining a consistent, extensible foundation. It uses a medallion architecture, where raw data (bronze) is preserved, standardized (silver), and then surfaced through clear precedence rules (gold). This allows multiple sources to coexist while maintaining traceability and data hygiene. It also builds a system working backwards from a future of worker voices and provider-submitted information and forwards from validated, but flawed, public datasets.

Programs, providers, occupations, and wages are connected through a matching system that is responsive to the catalog itself. As new records are introduced or existing ones change, relationships are re-evaluated. This allows the system to scale across states and providers and preferences in standardization methodologies.

The data pipelines are reusable and templatized - aka, they work in one shot. New datasets can be ingested into the same structure and participate in the same matching and prioritization logic. This supports alignment with frameworks like Advance CTE, enrichment from sources like the Harvard Workforce Almanac, and future layers like claimed programs or worker reviews, without any restructuring of the system.

Behavioral design

The experience adapts to the specific catalog it is built on. The quiz and discovery flows are dynamic, shaped by the programs, occupations, and data available in a given state instance. The tool puts a high premium on progressive disclosure, careful introduction of vocabulary, and above all - removing a lot of things that distract them from the path they came to walk. It's a short and yellow brick road right now.

The system is designed to help a worker move from exploration to decision. Information is structured so it can be understood and compared without translation. The experience surfaces relevant options early, then allows users to go deeper as needed. The goal is clarity in how to use it.

Built for partnership

The system is portable and designed to be deployed quickly across regions, but it is not fixed. Each instance can be shaped to a state or partner’s priorities, whether that is aligning to a specific framework, highlighting certain industries, or incorporating new data sources.

It is also designed to be used in partnership with stakeholders and workers. The structure allows for ongoing input, including program claims, updates, and feedback tied to specific records. This reduces the distance between the data and the people it represents.

In practice, this is what makes it different. Having a portable, scalable foundation (that can adapt to a state’s context, support different use cases, and improve over time while staying grounded in consistent data and real-world input) may be what's needed to land the adoption and motivation game in increasing worker enrollment in training programs across the U.S.

Dramatic, I know, but this is why I love my job! We've observed millions of workers interact with tools like these and know for certain that the opportunities are there.