Before I joined the product team at Motivity, I was a practicing BCBA® and then a customer success manager. In both roles, session notes were a constant topic. I coached staff on narrative writing, helped practices redesign their templates, and fielded calls about preventing notes getting kicked back by funders. Narrative writing is a complex skill to teach, and translating what happened into clinical language gets harder to articulate when your team has been in sessions all day. All while funder requirements keep stacking up.
When we started building Motivity’s AI narrative generator for session notes, we'd all watched documentation eat up hours that could be spent with learners. We’d also seen how differently each practice handles session notes, and our design had to be just as customizable.
What I saw helping practices with their session notes
The same challenges came up over and over. A supervisor would flag that many of the team's notes read almost identically, session after session. Or a funder would push back on a note for missing clinical details, and the RBT® who wrote it didn't know what was missing because nobody had spelled out what to include.
The hardest part was always the narrative. Structured fields like checkboxes or dropdowns were easy for people to fill out. But the open text box where you describe the session in clinical language? That's where staff spent the most time and had the most trouble.
Some staff wrote a few vague sentences every time. Others wrote too much, but none of it specific or objective. And the ones who wrote well took longer, because they were doing the mental work of turning observations into professional prose.
These patterns shaped how we approached the design of Motivity’s AI session notes.
The decisions we made and who we made them for
I want to walk through a few of the design choices behind Motivity's AI session notes and who we were thinking about when we made them.
The new RBT who needs guidance
A technician in her first few months doesn't always know what to include in a narrative. They might know the session went well, but “it went well” isn't an adequate session note.
We built internal-only fields for this. They can be used to create structured sections: dropdowns, checkboxes, and short-answer fields, that walk the author through documenting the session. How did the learner arrive? What interventions were used? Were there changes in routine? The AI reads those fields and turns them into narrative prose. Internal-only fields don't appear in the exported note, so the final document comes out with a clean narrative. But the supervisor can see what the author entered, allowing for more specific coaching.
This approach allows for guided documentation where the template does part of the teaching.
The experienced BCBA who doesn't need hand-holding
A behavior analyst writing a caregiver training note or a protocol modification typically knows what to cover. They may just need a place to type short-hand observations and have the AI translate into full sentences using clinical language.
For that, the same feature works with a simpler setup. The note author prompt area serves for jotting down notes and general guidance, while the AI composes from what the author gives it.
Multiple narrative sections
A protocol modification note might have two distinct parts: what you changed in the protocol and what direction you gave the technician. These are different topics that may warrant their own narratives for cleaner documentation.
In Motivity, AI is enabled per text box. You can have two narrative sections with two separate prompts, so each one is focused. One generates the protocol modification summary. The other generates the technician directions. The prompts stay narrow, and the output stays relevant to each section.
The practice where not everyone should use AI yet
Maybe your senior RBTs are ready but your newer staff need more supervised writing practice first. Or perhaps you want to pilot with one team before going wider.
Like most configurations in Motivity, AI access is controlled through a role-based permission. Turn it off for a role, and the same template works as a regular text box without an “AI” button. You don't need to create and maintain separate templates for staff with different levels of readiness.
Implementation mattered more than features
Before launch, a small group of provider organizations used the feature in their real workflows for about a month.
All of their templates were different. One organization built multi-field templates with detailed internal sections. Another had their BCBAs using structured tables and clinical shorthand. A third relied heavily on dropdowns. A fourth had staff inputting longer observations in a more transcript-style format.
That range was a design goal from the beginning, and seeing it confirmed across very different organizations was the moment I knew we'd gotten that part right.
Where experiences diverged were in prompt configuration. The organizations that spent time on their prompt instructions — specifying the session type, the structure they wanted, the terms to use — reported less editing per note and fewer complaints about length or terminology in the drafts. The organizations that enabled it without much configuration ended up doing that work note by note, writing longer observations and regenerating until the output matched what they wanted.
Three steps before you roll out
- Start with your existing templates. You don't need to rebuild from scratch. Pick one session type, enable AI on one text box, and add your prompt instructions. Our prompt bank has starting points organized by CPT code (97153, 97155, 97156, 97151) that you can copy and adjust. Starting from a blank page is intimidating even for people who are comfortable with AI. Take what's there and adjust it for your practice.
- Use the role permission to control your rollout. Enable AI for a small group first. Let your supervisors review those notes with the session data open. See what the output looks like with your templates, your learners, your session types. Then expand when you're ready.
- Read the first few drafts with your team. Not just to review them, but to talk about what's in them. What did the AI get right? What did it miss? What did the author need to add? Those conversations teach your team how to use this well, and they show you where your templates need adjusting.
If you're already using Motivity, your Customer Success Manager can walk you through template configuration and prompt writing. If you're exploring whether this fits your practice, talk to our team—we're happy to walk you through it.
