Fifteen minutes into a phone call, the whole plan changed.
The classroom data said attention-seeking — a 10th grader, disruptive whenever whole-class instruction ran long, worse whenever adult attention dropped or praise felt exaggerated. A reasonable read of the data. Then a parent picked up the phone and mentioned two things nobody at school could’ve known: a rough stretch of sleep and routine changes at home, and the one thing that actually calmed the kid down wasn’t a sticker chart. It was independent drawing and reading time.
That call didn’t confirm the plan already in progress. It rebuilt it.
Here’s the actual process that call belongs to — in order, built to find the real function of a behavior before anyone writes a strategy to change it.
Reframe Before Anything Else
Start by replacing “what’s wrong with this student” with “what’s this behavior getting them, and what’s it costing them to not have a better way to get it.”
Nearly every disruptive behavior serves a function: attention, escape, access to something preferred, or sensory regulation. Skip the hypothesis and write straight to consequences, and the plan ends up managing symptoms instead of the actual driver. It shows up again in a few weeks, wearing a different behavior.
Watch for the trap hiding inside good intentions, too. Exaggerated praise can accidentally reinforce the exact disruption it’s meant to redirect — a class laughing, an adult overreacting with attention, is still attention.
Step 1 — Collect ABC Data
Observe the student at least twice, in the setting where the behavior shows up most. Every time, record three things. The Antecedent — what happened immediately before. The Behavior — specific and observable, not “acted out.” The Consequence — what happened right after, including what the student gained or escaped.
Two observations is a floor, not a target. The more instances on paper, the more a real pattern shows up instead of noise from one bad Tuesday.
Step 2 — Call Someone Who Sees What You Can’t
This is the step from the phone call above, and it’s the step that gets skipped constantly because it feels slower than just writing the plan.
Before anything goes on paper, talk to at least one person outside the classroom — a parent, another teacher, the student. A short conversation can surface setting events that are completely invisible at school: a change in sleep, a shift in routine, something happening at home that reframes an entire week of ABC data. It can also surface reinforcers that actually work, instead of guessing.
In that one case, the token system already running at school wasn’t what motivated the student at all. Independent drawing and reading time was — a detail the classroom data never could have produced on its own, because it doesn’t happen at school.
Step 3 — Write the Plan
At minimum, the plan needs three things: the behavior impeding learning, the desired replacement behavior, and prevention plus response strategies.
The replacement behavior has to serve the same function the original one was serving. If the function is escape, the replacement is a legitimate way to request a break — not a rule against leaving the room, which just removes the exit without replacing the need.
Favor proactive strategies over reactive ones: pre-correction, one-step directions instead of multi-step, movement or choice breaks built into the schedule before the long instructional stretch starts — not tacked on after the behavior already happened.
Step 4 — Implement, Then Prove It Worked
Collect at least three data points after the plan starts, same setting as the original observations, and compare them directly against baseline.
This is the step that turns “we tried something” into “we know if it worked.” Skip it, and a plan can sit in place for a full semester without anyone actually confirming it’s doing anything at all.
Step 5 — Build the Family System Before the Meeting Ends
A plan that lives only at school drifts from one that lives at home within a few weeks — different language, different expectations, different follow-through.
Before the meeting closes, agree on three things. A one-page, family-friendly summary — target behavior, replacement behavior, a reinforcement menu everyone recognizes. A shared tracker that’s genuinely low-friction — a daily checkbox sheet beats a shared spreadsheet nobody opens. And a cadence: a daily tracker, a weekly summary, a longer check-in every several weeks, so the family isn’t only hearing from you on the hard days.
Precision Beats Assumptions
“I think this is attention-seeking” without ABC data behind it is a guess wearing a document’s clothes. Get the function wrong, and every strategy built on top of it underperforms, no matter how well-intentioned it is.
The unglamorous parts — the data collection, the phone call that feels like it could wait — are the parts that actually determine whether the rest of the plan works. Fifteen minutes on the phone is cheaper than a semester spent managing the wrong problem.