Built by operators who ran ABA clinics.

Collect the data once. Use it everywhere.

Q: What is ABA data collection software?

ABA data collection software is the tool RBTs and BCBAs use to record clinical data during Applied Behavior Analysis sessions — trial-by-trial responding, frequency and duration of behaviors, interval recording, task analysis, and ABC data — then graph it, track mastery, and use it to drive treatment decisions. In an all-in-one platform like Wilma, that same data also drafts session notes and feeds billing, so it is captured once and reused everywhere.

Q: What data types does Wilma support?

Wilma supports the full ABA data set: discrete-trial (trial-by-trial), frequency, rate, duration, latency, whole/partial/momentary interval recording, task analysis with step-level prompting, percent-correct, and structured ABC data. Targets and programs flow from the treatment plan so the data form matches the protocol.

Q: Does Wilma work offline?

Yes. Data collection is offline-first — RBTs capture full sessions on phone or tablet with no connection (home therapy rooms, rural drives, clinic dead zones) and everything auto-syncs the moment a connection returns. No data is lost.

Q: Does the data automatically graph?

Yes. Data graphs in real time as it is collected — cumulative records, rate graphs, trend lines, phase-change annotations, and per-target mastery. There is no export step and no warehouse delay. Wilma also proactively surfaces trends (for example a tact rate declining over three weeks) for the BCBA before the next session.

Q: How does data collection connect to session notes?

With the AI add-on (+$20/user/month, layered on Core $30/user/month), Wilma drafts the session note directly from the data the RBT just collected — goals linked, progress captured, payer-aware language. The BCBA reviews and signs. Because the data is captured once, it also feeds billing and progress reporting, eliminating duplicate entry.

Q: Is AI involved, and is it safe for clinical data?

AI drafts — humans decide. Wilma’s agentic AI turns collected data into note drafts and flags documentation gaps, but every clinical output requires BCBA review and sign-off before it is finalized or submitted. The underlying data is always the clinician’s; AI never changes a recorded data point.

Q: How much does it cost?

Core is $30/user/month and includes full data collection, graphing, mastery tracking, and supervision logging. The AI add-on (+$20/user/month) layers data-grounded note drafting and documentation-gap detection. Most clinical teams run Core + AI ($50/user/month). Core has a 10-licence minimum, so accounts start at $300/month.

Q: Can I migrate existing client data and graphs?

Yes. Active programs, historical data, and graphs transfer during onboarding so your BCBAs keep continuity of clinical decision-making from day one.

Q: How does Wilma compare to standalone data collection tools like Motivity?

Standalone clinical tools collect data well but then require bolt-on scheduling, billing, and documentation systems that don’t share a database. Wilma is one platform — data collection, notes, scheduling, billing, and phone on a single record. See the Motivity comparison.

Offline-first capture for every ABA data type. Real-time graphs and mastery. And the same data drafts the session note and feeds billing — so nobody enters it twice.

data capture, offline-ready: Real-time
progress + trends: Auto-graphed
session data → drafted notes: Note-ready
human approval on every output: BCBA-signed

What is ABA data collection software?

The clinical record for ABA — captured in-session, used everywhere.

ABA data collection software is what RBTs and BCBAs use to record clinical data during Applied Behavior Analysis sessions — trial-by-trial responding, frequency and duration of target behaviors, interval recording, task analysis, and ABC data — then graph it, track mastery, and drive treatment decisions from it.

The difference between a good and a great system is what happens to the data after capture. In a standalone tool, the data sits in a clinical silo and your team re-enters it for notes and billing. In an all-in-one platform like Wilma®, the data is captured once and reused everywhere: she drafts the session note, feeds the claim, and surfaces trends for the BCBA — with human sign-off on every clinical output.

Every ABA data type. Captured the way clinicians actually work.

Trial-by-trial, frequency, duration, rate, interval (whole, partial, momentary), task analysis, ABC, and percent-correct — collected on the tools your RBTs already carry. Targets and programs flow from the treatment plan, so data collection matches the protocol, not a generic form.

Discrete trial + natural environment

DTT and NET capture, switchable mid-session without losing context.

Behavior reduction

Frequency, duration, rate, and interval recording for target behaviors.

Task analysis & chaining

Step-level prompting data with mastery thresholds per step.

ABC & antecedent data

Structured ABC capture that links to the behavior intervention plan.

Works in the home, the clinic, and the dead zone.

RBTs collect on phone or tablet, online or off. Data syncs the moment a connection returns — nothing lost in a basement therapy room or a rural drive. The interface is built for one-handed capture during an active session, not for a desk.

Offline-first capture

Full session data collected with no signal; auto-syncs on reconnect.

One-handed UX

Large targets, fast switching — designed for live sessions, not forms.

In-session BIP support

An RBT can pull the agreed plan steps mid-session; the BCBA is notified.

Cross-device continuity

Start on a tablet, finish on a phone — one continuous session record.

Real-time graphs. Trends surfaced before the team meeting.

Data graphs the instant it is collected — cumulative records, trend lines, phase-change annotations, and mastery tracking per target. Wilma surfaces patterns (a tact rate trending down over three weeks) for the BCBA before the next scheduled session, not at Friday’s review.

Live progress graphs

Cumulative and rate graphs update as data lands. No export, no delay.

Mastery tracking

Progress tracked against the mastery criteria you set; you advance targets when ready.

Phase-change annotations

Condition lines and notes overlay the graph for clean visual analysis.

Trend detection

Wilma flags declining or stalled trends for BCBA review proactively.

Session data becomes a drafted note. The BCBA signs.

With the AI add-on (+$20/user/mo, layered on Core), Wilma drafts the session note from the real-time data the RBT just collected — goals linked, progress captured, payer-aware language. The BCBA reads it, adds a sentence, signs. Six minutes, not an hour. Nothing ships without human sign-off.

Data-grounded drafts

Notes generated from actual collected data, not free-text recall.

Goal linkage

Each note ties to the treatment-plan goals the session targeted.

Documentation gap flags

Missing data or signatures surfaced before the note is finalized.

Human approval, always

AI drafts; the BCBA reviews and signs every note. No exceptions.

Coach against the data. Supervision hours tracked to BACB.

BCBAs review the session’s collected data, leave coaching notes against it, and keep an audit-ready supervision trail in one place. Logged supervision is counted toward BACB requirements automatically — no calendar tetris, no paper trail.

Review against the data

Open the session’s data, review it, and coach against it.

Audit-ready trail

Supervision notes and attestation kept audit-ready.

BACB hour tracking

Logged hours and ratios counted toward BACB requirements automatically.

Coaching against data

Leave feedback against the session’s data.

Audit-ready data. Every access logged.

Every data point is timestamped and attributed. Full audit trail on PHI access, role-based permissions, and the documentation completeness checks that keep notes matching what each payer requires — caught before claims ship, not 90 days later.

Timestamped, attributed data

Who collected what, when — a defensible record for any audit.

Payer-aware completeness

Notes checked against payer-specific documentation requirements.

Role-based access

RBT, BCBA, biller, and owner each see exactly what they should.

HIPAA-grade logging

AES-256 at rest, TLS 1.2+ in transit, full PHI access audit trail.

Buyer's Checklist

Evaluating ABA data collection software? Demand every line.

FAQ

Questions clinicians ask about ABA data collection.

What is ABA data collection software?

What data types does Wilma support?

Does Wilma work offline?

Does the data automatically graph?

How does data collection connect to session notes?

Is AI involved, and is it safe for clinical data?

How is supervision tracked?

How much does it cost?

Can I migrate existing client data and graphs?

How does Wilma compare to standalone data collection tools like Motivity?

Is Wilma HIPAA compliant?

Customer Story · Clinical

Why a clinical team left "click-after-click" software behind

How fewer clicks for notes, data, and supervision gave a BCBA their day back.

See data collection that drafts the note for you.

Thirty minutes. Bring a real program and a recent session. We’ll show you the data captured, graphed, and turned into a signable note — live.

Choosing a system

How to evaluate ABA data collection software

The test of a data collection system is not what it can record. It is what an RBT can record accurately while also running the session — because the measurement your BCBA prescribed and the measurement your staff can actually execute have to be the same thing.

1Does it run the measurement systems your programs actually specify?Frequency and percent-correct are table stakes. The question is whether it also runs duration, latency, interresponse time, partial and whole interval recording, and momentary time sampling — with the interval timer on the same screen as the scoring control. When the prescribed method is slower than an approximation, staff quietly downgrade and the program document stops describing what is happening.Red flag: Interval recording that requires a separate timer, or duration that means typing a start and end time after the fact.
2Does it record prompt level, not just correct and incorrect?Percent correct is uninterpretable without prompt level — ninety percent correct under full physical prompting is not the same skill as ninety percent independent. A system that captures only accuracy forces the BCBA to reconstruct prompt level from memory at review, which is how programs advance on data that does not support it.Red flag: Prompt level captured as a free-text note rather than a structured field you can graph.
3What happens when the network drops?In-home sessions happen in basements and in cars. If the app requires connectivity, data gets written on paper and entered later — which is recollection, not measurement, and it is where interobserver agreement goes to die. Offline capture with reliable sync is not a nice-to-have in this field.Red flag: Offline mode that queues data but cannot start a new session, or that silently discards a session on a sync conflict.
4Does the data reach the note and the claim without being re-typed?Re-keying is where the error rate lives and where the hours go. Data collected in the session should populate the note's objective section and reconcile against the units billed. If those are three separate entries, you are paying for the same information three times and creating two opportunities for them to disagree.Red flag: A note template that starts blank, or billing that takes its units from a separate timesheet rather than the session record.
5Can a task analysis be scored step by step?Chained skills are scored per step or not usefully at all. A handwashing program recorded as a single sixty-percent figure hides whether the learner fails one step consistently — which means the grain size is wrong — or everything after step eight, which means the chain exceeds their stamina. Those need different responses.Red flag: Chains stored as one target with a percentage, with the step detail living in a comment field.

Before you sign

Questions worth asking on the demo

Show me an RBT running a duration measure and a frequency count in the same session without leaving the session screen.
How is prompt level stored, and can I graph accuracy against prompt level over time?
What exactly happens to a session started offline if the device does not reconnect for two days?
Does collected data populate the session note automatically, and can the clinician edit before signing?
How are task analyzes scored, and can I see the per-step pattern rather than a single percentage?
Can two observers collect independently on the same session for interobserver agreement?

Glossary

The terms you will hear

Continuous measurement: Recording every occurrence of a behavior — frequency, rate, duration, latency, interresponse time. More accurate than sampling and more demanding to run.
Partial interval recording: Scoring an interval if the behavior occurred at any point during it. Systematically overestimates how much the behavior is occurring, and more so as intervals lengthen.
Momentary time sampling: Scoring only whether the behavior is occurring at the instant an interval ends. Generally the least biased sampling method and the most practical when the observer is also teaching.
Prompt level: How much assistance a correct response required — independent, gestural, model, partial or full physical. Must be recorded alongside accuracy or the accuracy figure cannot be interpreted.
Interobserver agreement: Two observers recording the same session independently, compared to establish that the data describes the learner rather than the observer. Commonly held to an 80% floor with 90% preferred.
Task analysis: A multi-step skill broken into its component steps for teaching and scoring. Scored step by step, because the pattern of failing steps is what tells you what to change.

Keep reading

Related guides on running an ABA practice — the same operation, from a different angle.