Rich Henderson is Vice President of AI Solutions at Panorama Education, a K-12 platform for student support and learning, where he helps school districts bring data and AI together to improve student outcomes. As an influential K-12 education technology expert, he has spent the past decade focused on emerging technologies in K-12 and brings decades of experience directing emergent solutions for schools around the world. His background in special education, including early work as a special education assistant in Wake County Schools in North Carolina and with the Autism Society of North Carolina, shapes how he bridges the gap between cutting-edge innovation and practical classroom implementation. Rich guides districts through digital transformation, with a focus on the responsible and ethical use of AI. He advises education leaders on building sustainable AI strategies and developing essential AI literacy for staff and students. His work centers on a simple idea: technology should reduce administrative burden so educators can spend more time supporting students.
The federal IDEA child count deadline has passed for many districts across the country, but for special education leaders, the work is far from over. Staffing was set over the summer, and now caseloads are growing faster than the budget or existing teams can support.
I hear this challenge regularly in my conversations with district leaders. One leader in Texas recently told me their district added about 400 special education students in the first seven weeks of school. Now that school has already started, that means adding to caseloads of existing staff members or finding new special education resources.
And it’s not as simple as hiring more people. Special education positions are already hard to fill, especially once the school year is underway. Even when districts do bring on new staff mid-year, those educators are often learning new systems and processes while managing full caseloads. So, how can districts keep up with growing demand when adding more staff isn’t always an option?
I think the answer starts with treating the child count as more than a compliance report. It’s an opportunity to figure out where your team is stretched thin and which practices and tools could give them time back, without sacrificing the quality of support students receive. Leaders can take practical steps to addressing the child count data received.
Start by understanding what’s taking up educators’ time
Ask your team where their hours go after the count, and you’ll likely hear the same answer: gathering information.
Before anyone can decide what a student needs, they have to find out where that student stands. Who is making progress? What are the warning signs? Answering these questions often means pulling data from several systems, a process that gets repeated every time staff prepare for an evaluation, individualized education program (IEP) meeting, and Section 504 meeting. Practitioners already record their work. The information just sits in separate systems until someone has to piece it together.
It isn’t only a time problem. Students rarely face just one issue. Grades, assessments, attendance, and behavior all affect each other, and looking at them separately can hide the pattern that matters most. When that information comes together in one place, educators can recognize patterns more easily, better understand what may be contributing to a student’s challenges, and make more informed decisions about the support they need. That’s why a good first step is to understand how much time goes into assembling information, and find ways to bring it together so educators spend less time on data collection and more time on direct student support.
Standardize the playbook, individualize the plan
Individualized support doesn’t mean every plan or strategy needs to be built from scratch. Educators can personalize support for each student without starting over every time, which matters even more when caseloads are growing.
Instead, districts should establish a shared set of research-backed strategies and processes that educators can draw from based on each student’s needs. When a student faces a particular challenge, teams can select appropriate strategies at tier one, two, or three, monitor progress and adjust as needed. This helps educators develop high-quality, individualized plans without reinventing the process each time. It also gives newer staff a clear starting point, which is especially important when mid-year hires are stepping into full caseloads.
Requirements differ by state and district, so these shared processes need to align with federal, state, and local requirements before they can be customized. By establishing clear guidelines while leaving room for educators to tailor support to each student, districts can make the work more manageable without taking away the individualization that makes it effective.
AI can speed the work, but people remain in charge
With caseloads rising, it’s no surprise many special education leaders are turning to AI. It can give time back, but only when it’s used with care. Here are the risks I see most often and how to guard against each.
Quality and compliance: A draft can look so polished and thorough that it goes straight into the special education record. But generic AI that wasn’t built for education can produce work that isn’t tied to district or state requirements. Choose tools designed for special education, and make sure they work from your district’s own requirements for cybersecurity to protect and keep student data private.
Human judgment: AI can gather information, draft documents, and spot patterns, but people still decide the right services, present levels of performance, and goals. Treat every draft as a starting point. Real review means editing and questioning the draft, because moving quickly can mean missing something important.
Consistency: If you run the same student information through a process three times, you should get consistent results. Test any tool with sample cases before you roll it out. If the results vary, you can’t standardize or scale it.
Long-term reliability: Teams that build custom assistants inside consumer AI products can lose that work when a vendor changes its plans. Look for tools your district controls, and keep your playbooks and templates in your own hands so one product change doesn’t undo months of work.
Put these guardrails in writing. Many districts have a policy on student use of AI, but few address how teachers, special education leaders, and psychologists use it. A special education AI policy should spell out protections against bias and require that every plan be specific to the student. It should also define a review process for AI-generated drafts: who checks them against the district playbook and state requirements, what they look for, and who signs off, whether that’s an individual or the student’s care team. Clear steps mean fewer mistakes when caseloads climb.
Mesquite Independent School District in Texas, which serves more than 36,000 students, shows what using AI the right way looks like. Teachers there once spent two to three hours writing each IEP, and evaluation reports took experienced diagnosticians four to six hours and first-year staff up to 12, with much of that time spent gathering and re-entering data. To tackle that burden, the district used AI built for special education and grounded in its own student data to draft documents, while licensed professionals still review and finalize every one.
IEP drafting now takes about 45 minutes for some teachers, and evaluation reports are completed significantly faster. That gives staff more time for direct services and collaboration, and the district reports more consistent, higher-quality documentation across campuses. In other words, Mesquite paired AI with the safeguards above and got time back without giving up quality.
Three steps to take before winter break
With caseloads growing and resources stretched thin, special education teams don’t need to wait for a major overhaul to make meaningful improvements. There are three practical steps districts can take now to reduce administrative work, create more consistent processes, and help educators spend more time with students.
Map where the time goes. Ask staff how many hours they spend gathering student information before evaluations, IEP meetings, and Section 504 meetings. Then identify which systems to bring together so educators can see each student’s full picture in one place.
Build your shared playbook. Collect research-backed strategies by tier, check that they align with federal, state, and local requirements, and pair them with templates and training. That gives mid-year hires a clear starting point instead of learning the process by trial and error while their caseload grows.
Set your AI guardrails. Draft the special education AI policy and review the processes described above. Test any tool with sample cases before you roll it out, so that people make the final call on every plan.
See the student, then act
The child count tells you how many students you serve. It doesn’t tell you how to serve them well with the team you have. That comes from seeing each student’s full picture in one place, using a shared playbook so your team isn’t starting from scratch, and bringing in AI carefully, with people making the final call. Do that, and the child count becomes what it should be: a starting point for helping every student, even when the caseload outgrows the team.