Rad N Bad: Your Programming Sucks, Let's Make It Better — Individualized Goals, Layered Skills, and Generalization by Design
Presented by Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Many treatment plans are built by transferring items from a standardized assessment directly into goals, producing long lists of generic milestones that may not reflect the learner's life or family. In this episode, hosts Sean Yocum and Michael Carrero examine 'assembly-line' programming and offer alternatives grounded in behavior-analytic principles. They discuss writing fewer, more functional goals and layering prerequisite skills within a single meaningful target, rather than creating separate goals to fill authorized hours. They explain how teaching component skills in isolation on dense schedules can create prompt dependency and difficulty combining skills. They make the case that generalization is a design constraint rather than a later phase, drawing on Stokes and Baer's (1977) generalization strategies. Every goal should specify who must evoke the skill, where it must occur, and what will maintain it. Mastery should mean the skill occurs across people, places, and time under natural contingencies. Participants will leave with a goal-writing template.
Sign in to buy — $5One-time purchase, permanent access. Cert delivered to your inbox the moment you pass the quiz.
About this CEU
A milestone isn't a program, and an assessment isn't a curriculum. Mike and I take on copy-and-paste programming. We cover layering prerequisite skills into fewer, more meaningful goals, and why no goal should exist without a generalization plan. Real mastery means the skill survives without us. Disclosures & Publication Info No commercial products discussed. Both instructors are employed in ABA service delivery. Published: 2026-04-03 Listening to the podcast episode does not earn CEUs; credit is earned only by completing this OpenCEU event.
Abstract
Many treatment plans are built by transferring items from a standardized assessment directly into goals, producing long lists of generic milestones that may not reflect the learner's life or family. In this episode, hosts Sean Yocum and Michael Carrero examine 'assembly-line' programming and offer alternatives grounded in behavior-analytic principles. They discuss writing fewer, more functional goals and layering prerequisite skills within a single meaningful target, rather than creating separate goals to fill authorized hours. They explain how teaching component skills in isolation on dense schedules can create prompt dependency and difficulty combining skills. They make the case that generalization is a design constraint rather than a later phase, drawing on Stokes and Baer's (1977) generalization strategies. Every goal should specify who must evoke the skill, where it must occur, and what will maintain it. Mastery should mean the skill occurs across people, places, and time under natural contingencies. Participants will leave with a goal-writing template.
Credits
0.5 CEUs · 33 min
From the talk
What was covered
Fewer goals, prerequisite skills layered inside them, and a generalization plan for every target. Rad N Bad on programming and mastery.
- Justify every goal against a three, five, seven, and ten year outcome before it enters the plan.
- Write two layered goals instead of five split ones so prerequisite steps are taught inside a meaningful target.
- Stop giving each component skill its own dense reward schedule and then expecting the parts to combine.
- Document four things on every goal: who evokes it, where it occurs, what maintains it, and how your support fades.
- Call a skill mastered only when it holds across people, settings, and time on natural reinforcement.
- Plan caregiver reinforcement on day one, because a skill the home cannot maintain will not last.
Why Copy-and-Paste Programming Takes Over Caseloads
Most weak treatment plans do not come from weak clinicians. Mike points at response effort (how much work a task takes). Caseloads are full. A new assessment report lands. Goals are due this week. So the fast path wins. You open the template, spot a hole, and make the hole a target.
The hours math makes it worse. Teams act like there is a one to one ratio of hours to goals. More authorized hours means more goals on paper. So filler goals get written to close the gap. The plan gets longer. The child's life does not get better. Sean calls that assembly line therapy.
The cost is real. A list of generic steps ignores the messy, specific family living it. A checkbox is one slice of a skill area. It is not the whole area. When the slice becomes the goal, the plan loses its point.
But, you know, a milestone isn't a program and an assessment isn't a curriculum.
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Fix One: Build a Roadmap, Not a Six-Month Authorization
The first fix is the roadmap. Families need a clear, short plan for life. Not just a plan for this authorization period. Mike sits with the family and builds it with them. His test is simple. Can you say why this goal matters in three, five, seven, or ten years?
Most of us think too short. We teach the next thing in front of us. We rarely say what that skill unlocks later. Sean ties this back to Baer, Wolf, and Risley (1968). Behavior change counts when it lasts and shows up across settings. Generality (the skill shows up elsewhere) is a defining feature of the field. Yet plans still treat the future as someone else's job.
Parents are not shopping for targets. They want a trajectory. They want to see how this links to talking, independence, friendships, and safety. Sean also points to O'Connor (2010) on intensive behavioral intervention. Those outcomes were multi-domain and developmental. They were not one clean line of skills. So stop asking what you can teach by June. Ask what this child needs as a teen and an adult.
if you can't rationalize why you're putting a goal in for three, five, seven, ten years down the road, then it has no justification of being in that program.
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Fix Two: Teach Skill Domains, Not Assessment Check Marks
The second fix is teaching domains. A domain is a whole skill area, like communication. Sean admits the field chases clean graphs and percent mastered. Families do not live in percentages. No parent celebrates forty percent of goals met while their child still cannot use the toilet.
Standardized tools are not the villain here. The VB-MAPP, the ABLLS, PEAK, and the AFLS all have a job. Sean cites Padilla et al. (2021). Those tools were built to organize teaching in a developmental and functional order. They were never meant to be a list of random behaviors. Too often they get used to justify more targets instead of to guide the sequence.
Picture a learner who has mastered three hundred discrete targets (small, separate teaching items). That learner may still not ask for help. They may not handle frustration. They may not get through a school day without an adult at their elbow. That is not a data problem. That is a programming philosophy problem. Targets are the means. Domains are the end.
I think a score going up doesn't mean life is getting easier.
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Layering: Two Goals With the Prerequisites Inside Them
Here is the concrete move. You may not need five goals. Two may do the job. Inside those two, layer the prerequisite skills (the smaller steps that come first). Mike's example is sitting at a table. Step X comes first, then Y, then Z. All three live inside the table goal. Not every step needs its own data sheet.
Splitting them hurts the learner. Three goals get three sets of criteria. Each gets taught on a dense schedule. An FR1 schedule (a reward for every single response) builds a skill fast. Then you try to stack the three skills and thin the reward at the same time. The stack falls over. The child waits for help instead. That is prompt dependency (waiting on an adult cue).
The same logic applies to staff. Mike says a tech can hit eight targets inside two activities. Play a video game. Sing a song. Inside that you get choral responding, joint attention, and following a distal point (a point from far away). Track them line by line on paper. Run them together in real life. If staff cannot do that, it is a training issue, not a child issue.
Three separate goals that get three separate criteria that creates prompt dependency.
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Fix Three: Generalization Is a Design Constraint
Planning for carryover starts on day one. Mike borrows an army line: a plan is only good until it meets the enemy. Nothing stays perfect. If you wait for a perfect graph before you leave the room, you will wait forever. All that time you keep pumping the skill full of reward you never faded.
Then comes the real street. The child practiced at the fake crosswalk in the driveway. The real curb looks nothing like it. They run. The support was never thinned and the setting was never practiced. Mike is careful here. He is not telling you to start in traffic on day one. If a child bolts, safety comes first. Get the behavior manageable, then start practicing in the real spot.
Sean names the source. Stokes and Baer (1977) warned about train and hope (teach it and wish it spreads). That is a failure mode, not a strategy. He also points to Stokes et al. (1989). They named three tools. Programming common stimuli means using real life materials. Sufficient exemplars means using enough different teaching examples. Mediators are people or items that travel with the skill. Generalization (the skill spreads past training) is not a phase.
Why are they not successful in the environment? Did you take them to the environment?
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Redefining Mastery: Four Conditions, Not One Percentage
Now look at mastery criteria. The usual rule is eighty percent of opportunities across three sessions for four weeks. Hit the number and the goal moves to maintenance (checks run after the goal ends). Then it comes back on a random probe and the child misses it. Mike's point is simple. If it falls apart on the probe, it was never mastered.
Mastery means the reward has faded to a normal level. The natural result does the work. Ask a staff member where the bathroom is. The payoff is not praise for asking well. The payoff is getting to the bathroom. That is a mastered skill. A missed trial here and there is still fine. Nobody is perfect.
Sean lists four conditions. The skill happens across people, parents included. It survives a change of setting. It holds up over time. It contacts natural reinforcement (the real world payoff). Miss any one and the skill is still in training. A skill that only works with your materials, in your room, under ideal motivation is not mastered. As Sean puts it, that is performance art.
Because if I'm truly mastering a skill, I should be able to do it at home, in the clinic, at the grocery store, at my mom's, my dad's, my cousin, brother, you name it.
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
The Goal-Writing Template: Who, Where, What Maintains, How You Exit
Here is the template they leave you with. For every skill you teach, document four things. Who must evoke it. Where it must occur. What will maintain it. How your support comes out.
Start with who. Families hold most of the hours in a learner's week. Then where. A clinic only skill is a red flag. Name the home routines, the community spots, and the school settings. Sean's discharge test is blunt: if the skill cannot survive at Target or Walmart, it is not ready.
Next, ask what keeps the skill going without you. If the honest answer is that staff deliver the reward, the skill is still artificial. Last, write the exit. Spell out prompt fading, schedule thinning (fewer rewards over time), and staff withdrawal. The rule to live by is short. If you cannot explain how the skill survives without you present, do not teach it yet.
If the parent cannot evoke the skill, it's not generalized.
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Reinforce the Caregiver, Not Just the Learner
One piece gets skipped a lot. You have to reinforce the parent too. Mike can get a learner to do almost anything in session. If the home does not change with it, the skill dies there. So plan for caregiver behavior the way you plan for the child's.
Natural reward is usually just the thing that happens next. Flip the switch and the room has light. That is the payoff. Most people do not want a high five for it. Build that logic into the plan from day one.
Then use the three tier model you already have. Long term domains at the top. This period's goals in the middle. The teaching plan at the bottom. A communication domain is not a pile of labels. It can hold self advocacy and de-escalation (calming down and asking for space). Start where the learner's interest already lives. Mike says he has zero interest in the zoo, so why teach zoo animals first? The family steers the ship. Your job is to name the good stops along the way.
I don't need you to give me a high five for turning on the light.
From the talk — Sean Yocum, MA, BCBA, LBA & Michael Carrero, BCBA, LBA
Common questions
How many goals should a treatment plan have?▾
Mike's answer is fewer than most plans carry now. If five goals share prerequisite steps, two layered goals may cover the same ground. The steps still get taught, just inside one meaningful target. Goals written only to match authorized hours are filler.
What does it mean to layer prerequisite skills into one goal?▾
You pick the functional target first, such as sitting at a table for an activity. Then you list the steps that make it possible and teach them inside that goal. You do not create a separate goal and separate criteria for each step. That split is what builds prompt dependency and makes the parts hard to combine later.
When should generalization programming start?▾
Day one, not after mastery. Sean points to Stokes and Baer (1977) and their warning about train and hope. Waiting for perfect performance means the skill is loaded with reward you never faded. Safety is the one exception: get dangerous behavior manageable before practicing in a risky setting.
What should mastery criteria actually require?▾
Four things, per Sean. The skill occurs across people, including parents. It survives a change of setting. It maintains over time. It contacts natural reinforcement rather than staff delivered reward. A percentage hit in one room with one therapist does not meet that bar.
What four items belong on every goal?▾
Who must evoke the skill, where it must occur, what will maintain it naturally, and how your support fades. The fading section should name prompt fading, schedule thinning, and staff withdrawal. If you cannot explain how the skill survives without you in the room, the goal is not ready to teach.
About the speaker
Sean Yocum and Michael Carrero host the Rad N Bad podcast from Hickory Learning Group. In this episode they open their series on fixing weak practice by taking apart copy-and-paste programming. They speak from their own work writing plans, running intakes, and training session staff, and they include themselves in the critique.
This summary was generated from the recording’s transcript. Quotes are taken word for word from the talk.
What you'll learn
- 1Distinguish between assessment items or developmental milestones and individualized, functional treatment goals. Describe how to layer prerequisite skills within a single functional goal. Identify the generalization components every goal should specify: who evokes the skill, where it occurs, and what maintains it. Write mastery criteria that require performance across people, settings, and time under natural contingencies.
Course materials
Related talks
Talks that cover the same concepts
Rad N Bad: The Middle Ground of NET — Escaping the Babysitting Trap with Rigorous Natural Environment Teaching
Hickory Learning Group
Rad N Bad: The Dosage Dilemma — Contingency, Evidence, and Ethics in ABA Treatment Intensity
Hickory Learning Group
Rad N Bad: The Other Side of the Table — Ethical Failures, Compassionate Care, and Consumer Realities in ABA
Hickory Learning Group
Roadmapping: A Guide to Lasting Independence
Hickory Learning Group
Rad N Bad: DTT Is Not a Permanent Residence — Integrating Discrete-Trial and Naturalistic Teaching for Generalization
Hickory Learning Group
Rad N Bad: The Forbidden E-Word — Extinction, Its Side Effects, and Compassionate Implementation
Hickory Learning Group
More from Hickory Learning Group
Rad N Bad: The Myth of Assent? Assent as Behavior, Escape Contingencies, and Shaping Participation
Hickory Learning Group
So You Think You Can Lead, Ten Self Reflection Questions for Leaders in ABA
Hickory Learning Group
Rad N Bad: The Stereotypy Standoff — Autonomy, Function, and Contingency Design
Hickory Learning Group
OpenCEU is supported by advertising from third-party sponsors. Sponsors are clearly labeled, have no influence over course content, instructors, or CEU decisions, and never appear on certificates. The BACB does not sponsor, approve, or endorse OpenCEU sponsors or their products.