If a Behavior Plan Cannot Survive the School Day, It Was Never Finished
Updated: Oct 14
By Dr. Sarah Heller, Ed.D., BCBA, IBA, LBA-NY · Meaningful Metrics
It is 9:07 on a Tuesday morning.
The student's bus arrived late. The paraprofessional who usually supports arrival is absent. The behavior support plan says that the student should be prepared for transitions five minutes in advance using a visual schedule, offered a choice between two activities, and reinforced immediately after completing each step.
It is a thoughtful plan.
The visual schedule is in a binder across the room. No one prepared the choices because the schedule change was announced thirty seconds ago. The preferred reinforcer listed in the plan is not available, and the substitute paraprofessional has never been shown what implementation is supposed to look like.
By 9:10, the student is under a table, the rest of the class is waiting in the hallway, and a teacher is trying to manage a situation that the plan technically anticipated but never truly prepared anyone to handle. The unfortunate truth is that the bell does not wait for the plan to catch up.
I have seen versions of this moment unfold across schools, states, classrooms, and teams. The details change, but the pattern does not. A behavior plan is developed carefully. The recommendations are supported by thoughtfully and professionally completed assessment results. The procedures are described in detail and with care. The team reviews the plan and signs off on it. That same plan then enters a school day that was never designed to accommodate it, and the gap between what the plan assumed and what the environment could actually provide becomes visible in real time, with consistently sour results.
The question asked first when implementation begins to fall apart is almost always whether the staff followed the plan with fidelity. The question asked less often, and the one that matters more, is whether the plan was designed with enough understanding of the environment to make fidelity possible in the first place.
A plan lives in the hands of the people carrying it
A behavior plan exists during arrival, when several students need support at once, and during a fire drill, an unexpected assembly, a staffing shortage, a difficult academic task, and the five-minute transition after lunch when the classroom is louder than usual and everyone's regulation is already thinning. It exists when the teacher is absent, when the preferred item breaks, and when a strategy that worked beautifully last month no longer works in quite the same way and no one is entirely sure what changed.
A plan can be technically precise and still be functionally unusable. It can require more time than the classroom has, more staff than the building can provide, or more preparation than anyone was told would be necessary. It can depend on language that makes complete sense to the person who wrote it and no sense at all to the educator who has to interpret it while simultaneously teaching twenty other students. When implementation begins to look inconsistent or incomplete, it is easy to describe what is happening as a fidelity problem. While sometimes that description is accurate, other times it is a design problem that has been handed to the implementer to solve, and the burden of solving it falls on the people least supported to do so.
Contextual fit is not the softer side of clinical work
In behavior analysis, contextual fit refers to how well an intervention aligns with the values, skills, resources, routines, and supports of the people and environments responsible for carrying it out. It is sometimes treated as something to address after the technically correct plan has been developed, a secondary consideration for when time allows. That ordering misunderstands how behavior change actually works.
Benazzi, Horner, and Good (2006) examined what happened when behavior support plans were developed by school teams without behavior specialists, by behavior specialists working alone, and by school teams working with behavior specialists together. Plans developed by specialists alone scored highest on technical adequacy. Plans developed by teams without specialists demonstrated the strongest contextual fit. The plans developed through the collaborative condition, where behavioral expertise and knowledge of the student's actual environment were both present in the room, scored strongest on both dimensions simultaneously. No single condition produced both without the other. That finding is not a minor methodological footnote. It is a des
cription of what is lost when one kind of knowledge is treated as sufficient.
Monzalve and Horner (2021) took this further, examining technically adequate plans that were already in place but being implemented with low fidelity. Before any intervention on contextual fit, those plans were being implemented with fifteen percent fidelity, and student problem behavior was occurring during nearly half of every observation period. After teams reviewed the goals, identified where the plan was not fitting the environment, adapted procedures, and built clearer implementation supports, fidelity rose to eighty-three percent and problem behavior dropped to sixteen percent. The plan changed. The student changed. What did not change was the staff's commitment, because they had been committed the entire time. What changed was whether the plan they were asked to implement was one they could actually use.
The student is also a source of information about their own plan
This is where I want to add something the standard behavior support literature leaves out, and that a genuinely child-first and neuroaffirming approach requires.
A plan built entirely on what adults observe, report, and decide is a plan with a significant gap in it: the student. Including students who use AAC, who communicate through movement, through avoidance, or through the very behaviors that prompted the assessment in the first place, because they hold information about their own experience that no observation protocol fully captures. What feels overwhelming in a particular setting, what makes a transition harder or easier, which adults feel safe, and which strategies they are actually willing to use when things get genuinely difficult, as opposed to the ones they simply tolerate during low-stakes practice.
The growing literature on assent-based practice in ABA is explicit about this. When students have no genuine way to signal refusal or preference within an intervention, the plan teaches them that their experience of it does not factor into its design. That is not a neuroaffirming outcome, and it has real implications for whether the plan produces the kind of durable, meaningful behavior change it was created to support. Building student voice and preference into the design process, even when that requires creativity and flexibility in how that input is gathered, is not a departure from clinical rigor. It is an extension of it.
The fidelity question deserves more curiosity than it usually gets
Fidelity matters in our field, and it should. When an intervention is not implemented as intended, we cannot accurately evaluate whether the plan failed or was never truly attempted. But fidelity becomes a trap when it is used primarily to evaluate the behavior of the people implementing the plan rather than the plan itself.
Before interpreting incomplete implementation as resistance, unwillingness, or lack of competence, the team owes it to everyone involved to become genuinely curious about what the environment is communicating. Whether the staff received enough training to use the plan confidently and not just enough to recognize its components. Whether the steps can be completed while staff are simultaneously meeting their responsibilities to other students. Whether the materials are consistently available in the places they are actually needed. Whether the plan depends on one particular person being present. Whether anyone has asked the implementers which parts feel unclear, unrealistic, or impossible during the moments when support is most urgently needed.
A fidelity measure can tell us that only half of a plan's components were implemented consistently. It does not automatically tell us why, and the why is where the useful information lives.
What the Ethics Code already requires of us
Feasibility is sometimes discussed as though it is a concession made after the clinical work is finished. The Ethics Code for Behavior Analysts treats it as part of the clinical work itself.
Code 2.14 directs behavior analysts to consider stakeholder preference, implementation efficiency, and the likelihood that outcomes will maintain under natural conditions when designing behavior-change interventions. Code 2.16 asks us to explain the environmental conditions necessary for effective implementation. Code 2.19 requires us to identify and address environmental conditions that interfere with or prevent service delivery. These are not aspirational guidelines. They are professional responsibilities, and a recommendation that requires conditions the school cannot reliably create is not simply inconvenient. It may delay meaningful support, place unrealistic expectations on the people closest to the student, and leave a child experiencing the same distress while the adults around them debate why the plan is not working.
A plan that appears sound on paper but cannot be implemented consistently in the child's actual environment is not a clinical success. Feasibility is part of what makes the plan ethical.
The questions that belong in the design process, not after it
Questions about implementation should be part of how the plan is built, not a follow-up conversation after the document has already been written.
Before finalizing any recommendation, I want to know what the classroom looks like when the behavior is most likely to occur, including the pace of the room, who is usually present, what the teacher is trying to accomplish simultaneously, and what materials are realistically within reach. I want to know whether the person expected to collect data can do so without stepping away from instruction or compromising safety for other students. I want to know which parts of the plan depend on one person's memory or one material being available. I want to know how a substitute will understand what to do. I want to know whether the family recognizes their child in the document and whether the recommendations make sense outside the language of behavior analysis.
And I want to know, honestly, what the team will do on the days when the ideal conditions disappear, because those days are not exceptions. They are Tuesday in school environments.
The bell will ring whether the plan is ready or not
The school day will remain unpredictable. Students will arrive carrying experiences no one in the building knows about yet. The goal is not to design a plan that controls every variable. The goal is to design one that was built honestly enough to acknowledge the variables exist and realistically enough to hold together when they do.
That means building with the people who understand the environment, identifying the smallest essential actions that can be implemented consistently even on the hardest days, preparing for disruptions that are predictable even when their timing is not, and treating what the implementation team reports back as the data it actually is.
This is one of the conversations I will be bringing to the New York CEC Annual Conference in my presentation, "Stronger Together in the FBA: A BCBA-Educator Collaboration Framework," on Saturday, October 3, 2026, from 11:30 a.m. to 12:20 p.m. at Le Moyne College in Syracuse. The framework is built around a foundational understanding that a behavior plan is not finished when the recommendations are written. It is finished when the team has built something accurate, usable, and realistic enough to enter the actual life of the school, and when the student at the center of it has had a genuine hand in shaping what support looks like for them.
If you have implemented a behavior support plan as an educator, paraprofessional, clinician, family member, or administrator, I would love to know which parts of the process helped the plan fit your actual environment, and which parts expected the environment to fit the plan.
Frequently Asked Questions
What is treatment fidelity and why does it matter when a behavior plan is not working?
Treatment fidelity means the plan is being delivered the way it was designed, consistently and correctly by everyone who interacts with the student throughout the day. Research has found that even when evidence-based practices are adopted by schools, they are frequently not implemented with fidelity. When a plan is not working, the first question should be whether the plan has actually been tested, meaning implemented as written by everyone responsible for it, before anyone concludes that the intervention itself needs to change.
How do I know if the behavior plan is failing or if it is not being implemented correctly?
The distinction lies in the data. Aggregate behavior data tells you what is happening. Implementation data tells you why. If your team is only collecting information about whether the target behavior is occurring, and not about which components of the plan are being delivered by whom, you are missing the diagnostic information you need. A student who behaves very differently with different staff members is providing important information about implementation consistency, not necessarily about the intervention’s effectiveness.
Why is consistency across all staff members so important?
Behavior that works gets repeated. If even one or two adults in a student’s day are responding differently than the plan prescribes, providing attention when the plan says to withhold it or skipping a cue that is supposed to precede a transition, the student’s behavior is not failing to respond to the intervention. The student is responding perfectly rationally to an inconsistent environment. Research on barriers to BIP implementation found that inconsistent delivery across staff is one of the most prominent reasons plans fail in school settings.
When does a behavior plan actually need to be rewritten?
A plan genuinely needs revision when the function of the behavior was misidentified in the original FBA, when the replacement behavior being taught is not actually easier or more efficient for the student than the problem behavior, or when the reinforcers identified are not reinforcing in practice for that student in that setting. These are real possibilities and they are fixable. The key is ruling out the fidelity question first. Writing a new plan without addressing implementation gaps means the new plan will encounter the same problems as the old one.
What is the most common reason behavior intervention plans fail in schools?
Research consistently points to inconsistent implementation across the adults responsible for carrying out the plan. A statewide survey of over 600 teachers found that inconsistent delivery across staff was among the most prominent barriers to BIP implementation. Treatment fidelity is also rarely assessed in school settings, even though data shows that educators regularly struggle to implement interventions consistently. The plan is often not the problem. The support system around it frequently is.
Does training staff once on a behavior plan work?
Research comparing one-time training to ongoing coaching found that intensive, embedded coaching significantly increased educators’ fidelity to criterion levels, and that those gains transferred to real implementation with students in their actual classrooms. A single training meeting is rarely sufficient. Adult behavior, like student behavior, changes when there is ongoing feedback, practice, and support rather than a single informational event.
REFERENCES
Behavior Analyst Certification Board. (2020). Ethics code for behavior analysts. Updated August 2024. https://www.bacb.com/wp-content/uploads/2022/01/Ethics-Code-for-Behavior-Analysts-240830-a.pdf
Benazzi, L., Horner, R. H., & Good, R. H. (2006). Effects of behavior support team composition on the technical adequacy and contextual fit of behavior support plans. The Journal of Special Education, 40(3), 160–170.
Breaux, C., & Smith, S. (2023). Assent in applied behaviour analysis and positive behaviour support: Ethical considerations and practical recommendations. Journal of Applied Research in Intellectual Disabilities, 36(2), 238–248. https://pmc.ncbi.nlm.nih.gov/articles/PMC9897747/
Individuals with Disabilities Education Act, 20 U.S.C. § 1400 (2004).
Monzalve, M., & Horner, R. H. (2021). The impact of the Contextual Fit Enhancement Protocol on behavior support plan fidelity and student behavior. Behavioral Disorders, 46(4), 267–278. https://doi.org/10.1177/0198742920953497
Comments