The 6 Pillar Practices Every Special Educator Needs to Know (And Why Most EdTech Still Misses Them)

After a decade of research and a landmark 2024 update, special education has a clear answer to its hardest question: which practices matter most? Here are the six Pillar Practices, what they mean for your classroom, and why building technology to support them is harder than most think.

The Problem with “All 22 Are Important”

Special education has many frameworks, checklists, and professional development modules. Until recently, it lacked a clear hierarchy.

The Council for Exceptional Children (CEC) and the CEEDAR Center spent nearly a decade identifying and refining a core set of High-Leverage Practices (HLPs): 22 techniques frequently used, broadly applicable, and backed by research. First formalized in 2017, the framework was significantly updated in the 2024 second edition, introducing a change that may quietly reshape how we train teachers and build tools.

For the first time, the framework made a direct claim: not all practices are created equal.

The 2024 revision introduced a formal hierarchy: “Pillar” practices and “Embedded” practices. The six Pillars form the foundation of effective special education. Master these first, the framework argues, and the rest becomes possible. Skip them, and no programming, curriculum, or technology will close the gap.

So what are they, and what do they actually demand of a classroom, a team, and a platform?

The Six Pillar Practices, Explained

HLP 1: Collaborate with Professionals

No educator can fulfill an IEP alone. A student’s plan involves general educators, specialists, paraprofessionals, administrators, and related service providers. When these professionals work in integrated teams, service delivery becomes coherent and outcomes improve.

In practice, this means shared planning time, clear hand-offs of goal ownership, and a common language so the speech-language pathologist and reading specialist do not work at cross-purposes. Researcher Matthew Ronfeldt and colleagues found that teachers who participated more in high-quality collaborative structures produced better student achievement gains, even after controlling for other variables.

The friction is real: most schools lack built-in structures for this. Collaboration is squeezed into the margins of an overcrowded day.

HLP 3: Collaborate with Families

Family collaboration is not about sending home a quarterly progress report. Research frames it as “co-designing” the learning experience: families bring knowledge about their child that no evaluation can fully capture, and educators translate data into something parents can use.

This requires active reflexivity from educators: examining assumptions and biases about what families can contribute, especially when linguistic and cultural backgrounds differ from the dominant school culture. Evidence shows strong home-school partnerships are consistent predictors of student academic success and social-emotional adjustment.

It also requires systems. If families get information about their child’s progress only through formal IEP meetings twice a year, that is reporting, not collaboration.

HLP 6: Use Student Assessment Data, Analyze Instructional Practices, and Make Necessary Adjustments

This is the intellectual engine of the framework. The 2024 refresh rebranded the old “Assessment” domain as “Data-Driven Planning” — and that shift matters. Assessment is no longer a static event producing a document. It is a continuous feedback loop driving every instructional decision.

Systematic reviews show that when teachers use frequent progress monitoring data to decide whether to continue, modify, or intensify an intervention, learning gains are significantly higher than in traditional instruction. The key word is “pivot”: HLP 6 lets a teacher catch problems before a student spends months on the wrong path.

This practice also makes the rest of the Pillars coherent. Without data informing decisions, explicit instruction is just a structure. With data, it becomes a precision instrument.

HLP 7: Establish a Consistent, Organized, and Responsive Learning Environment

HLP 7 is the prerequisite for everything else. Not metaphorically but literally. No instructional technique, no matter how evidence-based, will gain traction where students don’t feel safe, routines are unpredictable, or behavioral expectations are unclear.

The research here is dense and consistent. Hattie’s synthesis of over 800 meta-analyses identified effective classroom management as one of the highest-leverage influences on student achievement. The mechanism is straightforward: a responsive environment maximizes academic learning time and minimizes the behavioral friction that pulls both the teacher’s and the student’s attention away from instruction.

One evidence-backed target within this pillar is a 4:1 ratio of positive, specific feedback to corrective statements. This is not feel-good pedagogy. It creates psychological conditions where students with disabilities — who often have histories of academic struggle and corrective feedback — feel safe enough to take learning risks.

HLP 16: Use Explicit Instruction

If there is a single practice most synonymous with effective special education, it is this one. Explicit instruction is arguably the most rigorously studied instructional technique in the field, supported by decades of meta-analytic evidence across subjects and disability categories.

The structure is precise: activate prior knowledge, model the skill clearly (“I do”), move to guided practice with scaffolding (“We do”), then release to independent practice as fluency builds (“You do”). Brophy, Rosenshine, and Hattie — the major syntheses — all converge here. Explicit instruction works because it reduces cognitive load, anticipates common errors, and provides feedback loops students need to build and consolidate skills.

It is worth naming why this matters especially for students with disabilities: many have processing, memory, or attention profiles that make incidental or discovery learning difficult. They do not pick up what other students absorb implicitly. Explicit instruction removes ambiguity.

HLP 20: Provide Intensive Instruction

This is the most specialized of the six Pillars and, in many ways, defines special education as a distinct professional field rather than a modification of general education.

Intensive instruction means dramatically increasing the frequency, duration, and precision of teaching for students not making progress with standard interventions. It operates within the Data-Based Individualization (DBI) model: the teacher uses frequent progress monitoring to decide whether to continue or change an approach, often working in very small groups or one-on-one.

Nelson and colleagues’ 2022 systematic review identified HLP 20 as having the strongest experimental evidence of any HLP. It is, in the most concrete sense, the highest-leverage practice in the framework and the one most likely compromised when caseloads grow, scheduling tightens, or inclusive models leave no time for pull-out support.

Why This Hierarchy Matters for Technology

At Lessi, we have spent much time thinking about what it means to build tools that actually support these practices, not just the easy ones.

Most educational technology platforms are built around a 1:1 model: one student, one device, and an adaptive algorithm that adjusts difficulty in real time. That model maps cleanly onto certain embedded HLPs, such as providing scaffolded supports (HLP 15) or offering assistive technology (HLP 19). And those things matter.

But look at what the six Pillars are actually asking for.

HLPs 1 and 3 are fundamentally about relationships and team coordination. They require shared visibility across professionals and families — not a dashboard one person looks at in isolation. HLP 6 requires that data from evaluations, progress monitoring, IEP goals, and daily instruction be connected in a way that allows a teacher to make a coherent decision. HLP 16 requires a teacher, not an algorithm, to make deliberate choices about modeling, guided practice, and feedback. HLP 20 requires the kind of clinical precision that only occurs when a professional actively interprets data and adjusts in real time, in a real relationship with a student.

When we were building the Lessi AI Classroom, this is where the hardest design work happened. It wasn’t the AI that was difficult to build; it was the hand-offs. Evaluations produce data. That data should inform IEP goals. Those goals should connect to curriculum choices. Daily instruction should generate progress data. That data should flow back to families in a form they can understand and act on. And all of it should close the loop back to the team.

We went through three full iterations before the architecture actually worked. Not because any single piece was technically complex, but because, in practice, each component produces its own data in its own format, on its own timeline, and is reviewed by different people with different roles. Connecting those moments without losing fidelity — and without creating more work for already-overwhelmed educators — turned out to be the real problem.

The 1:1 platform model doesn’t have this problem because it doesn’t try to support the system. It optimizes for the student’s screen time. That’s useful, but it leaves the most important HLPs entirely unsupported.

What This Means for How We Think About Teacher Development

The 2024 framework is also a roadmap for how we train and support teachers, not just how we build tools.

A consistent finding in the research is what some scholars call the “Knowing-Doing gap”: most special educators can accurately describe the HLPs, but systemic barriers — such as caseload size, limited planning time, and limited administrative support — prevent consistent implementation. A teacher’s struggle to implement often isn’t a knowledge problem. It’s a design problem.

The framework’s response is a push toward practice-based teacher education: more simulation, more supervised clinical experience, more coaching in the field. And a strategic emphasis on the six Pillars for novice teachers, because you can’t develop depth in all 22 simultaneously, and trying to leads to surface-level mastery of none of them.

For school leaders, the implication is clear: professional development that treats all 22 HLPs as equally urgent is likely less effective than one that builds deep fluency in the Pillars first, then uses that foundation to introduce the Embedded practices.

A Final Thought for Practitioners

The 2024 HLP Refresh is not another checklist to post in your team room. It’s a clarification of what the research has been trying to tell us for years: a small number of practices, executed with real skill and consistency, drive most of the outcomes we care about.

If you’re a special educator feeling overwhelmed by the breadth of what effective practice demands, this framework is actually good news. It gives you permission to prioritize.

Start with the environment (HLP 7). Build the explicit instructional routines (HLP 16). Make sure your data is actually driving your decisions (HLP 6). Get the team and the family genuinely in the loop (HLPs 1 and 3). And for the students who need more, intensify with precision (HLP 20).

Everything else builds on that foundation.

References

Council for Exceptional Children & CEEDAR Center. (2024). High-Leverage Practices for Students with Disabilities (2nd ed.). Council for Exceptional Children.

Hattie, J. (2008). Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating to Achievement. Routledge.

Nelson, G., et al. (2022). A systematic review of meta-analyses in special education: Exploring the evidence base for high-leverage practices. ResearchGate. https://www.researchgate.net/publication/357415022

Ronfeldt, M., Farmer, S. O., McQueen, K., & Grissom, J. A. (2015). Teacher collaboration in instructional teams and student achievement. American Educational Research Journal, 52(3), 475–514.

Rosenshine, B. (2012). Principles of instruction: Research-based strategies that all teachers should know. American Educator, 36(1), 12–19.