The education sector has no shortage of ideas for solving the literacy crisis: AI tutors, new interventions, personalized learning. Every few months, a new initiative promises to reverse America’s reading recession.
Yet today, 65% of 12th graders aren’t proficient in reading, according to the latest results on the National Assessment of Educational Progress, and the same tests show reading achievement has been declining for more than a decade. Before we add another tool to the initiative pile, we need to ask ourselves: What is the sum of all these “solutions,” and how are kids experiencing them?
I learned something important about this over 10 years ago while teaching reading to 7th graders in rural Mississippi. One of my students—I’ll call him Johnny—spent months slumped over a 3rd grade workbook, disengaged in my remedial English class. In early spring, he wandered into my honors class while we were reading The Diary of Anne Frank. Within minutes, he lit up. He participated, asked questions, and made connections using words I’d never heard in my remedial class. I realized I had made assumptions about his abilities, supported by a system that told me I was doing the right thing.
This system wasn’t designed to advance Johnny in reading. It was designed to “meet him where he was” by lowering our expectations of what he could do. While administrators could demonstrate we were “serving” every kid, we boxed Johnny into his struggle with low-rigor tools.
Since then, I’ve spent a decade working in education—observing classrooms, examining research, speaking with hundreds of teachers and administrators, and founding an education nonprofit—to understand why so many kids are never given the chance to become strong readers. I keep meeting Johnny, each time with a different name and a different intervention that asks less of him.
The New Teacher Project, an education policy nonprofit, found that students spend about 74% of their time on assignments that aren’t grade-appropriate. This is most likely to happen when every layer of a school system operates independently. Learning is cumulative. Every lesson should prepare students for the next, every grade should build on the one before it, and interventions should support rather than replace the grade-level material. Instead, students experience a patchwork of disconnected tools. During the 2024–25 school year, the average student used 48 different ed-tech tools.
Districts should begin with a coherent core curriculum that gives every student access to grade-level instruction.
It starts in the classroom. Like Johnny, struggling students are routinely pulled out of grade-level instruction and placed into interventions that often use different materials and instructional approaches. Support intended to accelerate learning instead keeps students where they are. It continues across grades.
In many schools, there’s no guarantee that what students learn in 7th grade builds on what they learned in 6th. Teachers work hard to build meaningful lessons but often in silos.
Zoom out to the systems level and you’ll sometimes see the lack of alignment between assessments and what’s being taught in the classroom. Recent research on Common Core-era state assessments found better alignment between standards and measured content than with past state tests, but even with the newer tests, gaps continue. Teachers are experiencing a similar challenge, often receiving professional development that doesn’t align with their curriculum or assessments.
Here’s the thing: We know students progress faster in coherent systems where each lesson builds on the last. When curriculum, instruction, intervention, and assessments reinforce one another, students make stronger academic gains. Research on Chicago schools first showed this in 2001, and a 2025 randomized trial in Tennessee confirmed it. Following a redesign of Knox County’s early-literacy intervention system for coherence, the lowest-performing students gained 1.8 additional months of learning per school year. For struggling students, the ones who have suffered the most in the reading recession, this additional learning is critical.
And yet, we continue to introduce solutions that compete with one another. Why?
Part of the problem is scale. No administrator can be in every classroom, every school, and every district to see all the redundancies and fragmentation. It’s invisible from any single point of view in the system.
But the harder truth is that coherence doesn’t sell. Vendor sales of instructional materials, professional development, and ed tech to school districts runs to tens of billions of dollars. Incentives favor point solutions that solve one issue (e.g., fluency or comprehension) rather than coherent, systems-level solutions that require change management and switching costs. A curriculum director may choose core materials, while a Title I coordinator an intervention software, and a superintendent a PD contract. The incentive at every step is to solve the problem in front of you, not to ask whether the solution works with the 10 other tools that already exist. Nothing in the way we buy, fund, or evaluate education tools rewards coherence, and there’s a lot that rewards the opposite.
Frankly, teachers and administrators shouldn’t have to stitch together coherence between products that weren’t designed with the whole system in mind. We need stronger upstream incentives and quality checks so tools that don’t fit the system aren’t even considered. Then, we need accountability so tools that don’t work get bagged before they do more damage.
It will take political will and coordination to reset these incentives. But I suspect the for-profit market would design better solutions if schools demanded coherence. State legislatures should require schools to evaluate whether supplemental programs and initiatives reinforce—or disrupt—the school’s core curriculum as a condition of funding.
In the meantime, there is a lot that schools can control to shape the demand. Instead of cobbling together dozens of supplemental tools, districts should begin with a coherent core curriculum that gives every student access to grade-level instruction. Every intervention, assessment, and professional development session should strengthen that foundation—not compete with it.
Kids spend nearly 15,000 hours in school. That is ample time to teach them to read and write.
How many Johnnys are we going to lose to a system of disconnected interventions? Before we adopt the next literacy initiative, let’s make sure the ones we have are working together.
