How to Ask RBTs for Feedback on Your Supervision (And Get Real Answers)

The two best moves you can make when you ask a Registered Behavior Technician (RBT) for feedback on your supervision are these: put a number on the ask ("tell me two things I can start or two things I can stop") and when they answer, say zero words of explanation. Just thank them and move on. That is the whole job. If you do those two things, the rest of this page is detail.

Most BCBAs do not get useful feedback from the RBTs they supervise because they ask the wrong question and then react the wrong way. They say "how am I doing?" and get "good!" and walk away thinking the relationship is healthy. It is not. They just gave the RBT an easy out. And on the rare day an RBT does volunteer something honest, the BCBA explains why the thing happened. The explanation feels normal. To the RBT, it feels like getting in trouble for telling the truth.

This page shows you how to fix both halves of that. The tactics are pulled from a BCBA-led CEU by Mellanie Page on supervision, and they are simple enough to use in your next session.

Why "how am I doing?" gets you nothing#

"How am I doing?" is a yes/no question dressed up as an open question. The brain hears it and reaches for the easiest socially safe answer. That answer is almost always "fine" or "good." Nothing about the question forces the RBT to actually search their memory for a specific thing you did or did not do.

There is also a power gap. You write their performance reviews. You sign off on their hours toward BCBA. You decide which cases they get. When you ask a wide open question, the RBT does a fast risk calculation: "Is it worth the awkwardness of saying something real?" The answer is usually no.

So the question itself is the first problem. You are not asking for feedback. You are asking for permission to feel okay about yourself. The RBT can read that, even if neither of you names it out loud.

The two-things rule: quantify the ask#

The fix is to put a number on the request. Ask for two things. Two things to start doing, or two things to stop doing. That is it.

Will you tell me two things I can start or two things I can stop that would improve our supervision? The reason that you do this is because the brain wants to fulfill that sort of request, that inventory request. From the talk — Mellanie Page

When you say "two things," the brain treats it like a checklist with two empty boxes. It does not want to leave boxes empty. So it searches. It comes up with something even when "nothing comes to mind" was the easy answer a second ago.

You can swap the number. Three things. One thing. The point is that any number closes the escape hatch. Open ended questions let people opt out. Quantified ones do not.

Use this in a real sentence with a real person. Try it at the end of your next one-on-one. Say it casually. Do not make it feel like a formal review. Then wait. Do not fill the silence. The silence is where the answer lives.

The zero-explanation reply (and why excuses punish honesty)#

Here is the harder part. When the RBT actually tells you something, your job is to say thank you. That is the whole reply. No "but." No "well." No backstory.

This is hard because it feels rude. You want to give context. You want them to know there was a reason your laptop was open during last week's session. The reason is good. The reason is true. Say it anyway and you have just trained the RBT to never give you real feedback again.

In behavior terms, your explanation is a punisher for the act of speaking up. The RBT took a risk. They told the truth. You followed it with a defense. The pairing teaches them that honest feedback leads to a small social cost. Next time they will skip the cost and say "everything's good."

If I start then explaining, oh, but they just had a reauthorization coming up... that person is immediately going to be averse to giving you future feedback. From the talk — Mellanie Page

So bite your tongue. Even when the explanation is airtight. Even when you are sure the RBT misread the situation. Especially then. Thank them for the feedback. Write it down. Move on.

If the feedback was wrong on the facts, you can come back to it later in a different conversation. But not in the moment they handed you the feedback. That moment is sacred. Mess it up once and the next four asks will get you nothing.

Showing change in the next session instead of defending yourself#

If you cannot defend yourself in the moment, what do you do with the urge? You spend it in the next session.

say thank you for the feedback, and just move on. And so if that truly was the case that you had a reauthorization coming, then your next supervision session should be banging. From the talk — Mellanie Page

The phrase "should be banging" is the whole rule. If an RBT said you were buried behind your screen last week, then next week your laptop stays closed. You are on the floor with them. You make eye contact. You comment on the work. You ask about the kid. The RBT notices. They do not need you to announce it. The change is the announcement.

This is just behavior change applied to yourself. You got feedback. You modified the behavior. Future feedback is more likely because the RBT learned that talking to you produces real change. That is reinforcement working in the direction you want.

Show, do not explain. Three words. Tape them to your monitor.

Anonymous surveys for the RBTs who won't say it to your face#

Some RBTs will never tell you the hard thing in person. Their personality, their history with bosses, or the size of the power gap will keep them quiet no matter how warm you are. You still need their input. So you build a back channel.

anonymous feedback surveys to staff, informal questions and feedback forms after every observation that the RBT can give us feedback as well. From the talk — Mellanie Page

The two pieces are anonymity and frequency. Anonymity removes the risk of being identified. Frequency normalizes the act of giving feedback so it does not feel like a big deal when something real comes up.

Use a free tool. Google Forms is fine. Keep it short. Three questions, maybe four. One number ("rate the last two weeks of supervision 1-5"), one open ("what is one thing I should start doing?"), one open ("what is one thing I should stop doing?"). Send it every other week. Or build it into your post-observation form, where the RBT rates you back after you rate their session.

The first few rounds will be bland. People test the system. Around the third or fourth round, when nothing bad has happened to anyone who gave honest input, the real answers start coming. Wait it out.

A weekly feedback ritual you can copy#

Pull all of this into one weekly habit. Here is a copy-paste version.

In your weekly one-on-one, end with one sentence: "Tell me two things I can start or two things I can stop that would improve our supervision."

When they answer, say "thank you, I appreciate that" and write it in the same notebook you use for case notes. No follow up question. No defense. If you must say one more thing, make it "I'll work on that."

Between sessions, do the thing. Pick whichever item is easiest to change first. Make the change visible.

Every two weeks, send an anonymous form. Three questions. Same questions every time so RBTs know what to expect.

Once a month, look back at what came in. Pick one pattern that has shown up more than once. Address it openly with the whole team. Do not name who said what. Just say "I've heard a few times that observations feel rushed. Starting next week I'm blocking an extra ten minutes for debriefs." Then do it.

That is the ritual. Five steps. None of them require any tool you do not already have.

Frequently asked questions#

What do I do if the feedback I get is genuinely wrong or unfair? Thank them in the moment. Write it down. Do not push back. Two or three days later, in a different conversation, you can say "I've been thinking about what you mentioned last week. I want to share the context, because I think we might be looking at the same situation differently." Now the feedback is settled history, not a live debate. You can talk about it like two adults solving a problem instead of one person defending and one person attacking.

Should I share what RBTs said about me with my own clinical director? Share patterns, not quotes. If three RBTs have told you in different ways that your observations feel rushed, that is a pattern worth raising with your director, because it might affect how you schedule your caseload. Quoting one RBT to your boss is a different thing. It can break trust fast, especially if the RBT figures out who said it. Use the surveys to spot trends. Talk to your director about trends. Keep individuals out of it unless there is a safety or ethics concern that requires it.

How long until anonymous surveys actually surface real feedback? About three to five rounds. The first one or two will get you "everything is great" or one liners. People are watching to see if anything bad happens to whoever speaks up. If nothing bad happens, the honest answers start arriving. Stay consistent. Do not skip a cycle because the early rounds feel pointless. The point is the consistency, not the early results.

Try it once this week#

Pick your next supervision session. End it with the two-things ask. When the answer comes, say thank you and shut up. Then change the thing by the following week. That is the whole loop, and it is the loop most BCBAs never run because they cannot get past the urge to explain.

If you want the rest of the playbook, watch the full talk. Mellanie walks through the surveys, the observation forms, and the rest of the ritual with examples from real supervision sessions. It is one supervision CEU and it pays for itself in the first week you use it.