Charlene Kurth on ABA software adoption: why practices buy tools before defining the problem, and the pre-mortem she runs before any rollout starts.
Key Takeaways
- ABA software adoption is measured by what happens when nobody is chasing anyone: Kurth’s test is whether a system keeps running with no designated caretaker, and whether a skipped step is felt by the person downstream. She spent years as that caretaker and describes the role as a trap that reads, from the inside, as indispensability.
- Most ABA software rollouts are decided before anyone logs in: The common failure is buying a product because a peer of similar size has one, then switching on every feature at once. The alternative sequence starts with a single defined problem and a configuration that solves only that.
- RBT turnover raises the stakes on ABA software usability: Median annual turnover among Registered Behavior Technicians runs near 65 percent by industry reporting, which makes every hire a training event on whatever system a practice runs. Physician research has separately graded electronic health record usability an F on a standardized scale, with poorer usability associated with higher odds of burnout.
- A pre-mortem surfaces the objections an implementation kickoff suppresses: Kurth opens implementations by asking a team to assume the project has already failed and explain why, a technique published in Harvard Business Review in 2007. What surfaces then gets sorted by likelihood and impact, before there is a timeline to defend.
At the ABA C.A.R.E.S. Summit in Boston in August, Charlene Kurth delivered a line from a vendor panel that sounded workshopped: if one person is quietly holding your software together, that is not adoption. Kurth, Chief Operating Officer at the practice management company Office Puzzle, said in an interview on September 9 that it had not been workshopped at all. “Not rehearsed at all,” she said. “It’s just such a pain point in my experience, because I was that person.”
The person she means is a familiar figure inside a growing practice. Early in her career, Kurth said, she was process oriented enough to volunteer for every implementation, and she built systems elaborate enough that she was the only one who kept them clean. She chased colleagues for entries. She ran meetings in which she asked, one by one, whether people had completed a step, then updated the system herself while they answered. “For a long time in my career, I was really proud of that,” she said, “because one, it makes you an indispensable person, in your mind. It’s like, they couldn’t run this without me. But it’s not what’s best for the company, because they can’t run it without you.” The second cost was her own calendar. “Your job becomes making sure that a system works instead of all the other things that you can do with your time.”
Her definition of the opposite is deliberately unglamorous. “Adoption is when the system runs smoothly without you,” she said. “It does not need a person.” What replaces the caretaker is not enforcement but sequence. The colleague downstream, who cannot finish her own work until an upstream step is done, is the one who notices the gap, which moves accountability into the workflow instead of onto somebody’s follow-up list. The test is whether an omission registers at all. “It has so much value that if it’s not done, it’s felt, and so it becomes second nature,” she said. Her example was ordinary rather than clinical: nobody at her company checks Slack because she asks them to.
Where ABA Software Rollouts Fail: Buying a Product Before Defining the Problem
Asked where implementations actually die, Kurth described two endings. In the first, the practice never commits. Stakeholders cannot agree on what they are solving, the search stalls, and the organization stays where it is, because, as she put it, the fear of what future pain could be is bigger than the pain that they know. In the second, the purchase goes through and nothing changes. Staff follow the new steps, the invoice arrives every month, and nobody can name the thing that got better. “Now you just have added steps with no value,” she said. She considers that ending the worse of the two, and thinks it is common because adding is easier than editing. “Everyone loves adding,” she said. “People rarely think about removing or editing out from their day.” She was one of five vendors on a panel that spent an hour telling buyers how to make a sales process harder to survive.
The root cause she names sits earlier than the rollout. “They might just say, ‘Oh, I’m at a certain size. I saw someone else has XYZ technology. I need to buy that technology,’” she said. “They’re buying a solution instead of first defining their problem.” The second version of the error arrives just after the contract, when a practice that has paid for a platform decides to switch on all of it. Her counsel is to implement narrowly, solve the defined problem, then expand. “You don’t want to buy something and be swayed because it has 10 other features that have nothing to do with your biggest challenges or pain points in your business.”
That advice reaches an audience with little margin for a wasted year. Median annual turnover among Registered Behavior Technicians runs near 65 percent by industry reporting, which makes every new hire a training event on whatever system a practice runs, and a 2024 University of Florida study of Florida technicians identified difficulty establishing competency as one of four recurring drivers of burnout and turnover. The usability problem is not unique to behavioral health. A 2020 study in Mayo Clinic Proceedings scored physician-rated electronic health record usability at 45.9 on the System Usability Scale, a grade of F against benchmarks drawn from other industries, and found each point of better usability associated with 3 percent lower odds of burnout. In a sector where the operational mistakes that break multi-site growth tend to be the ones nobody priced, a year spent inside a system nobody uses is a real number on a real income statement.
People, Process, and Technology: Diagnosing Which One Actually Broke
Kurth’s framework for diagnosis is older than the market she sells into. “People, process, and technology, they always go together,” she said, and the mistake she sees most often is treating one as separable from the others. A practice growing fast and losing track of itself has an alignment problem, which points to one class of answer. A larger practice finding gaps in billing and claims has a different problem and a different answer. The distinction gets harder to make in organizations where clinical and operational leadership have split without a rule about who owns what.
Sometimes the answer is not a purchase. “I know this goes against my interest, I’m selling a tool,” she said, before making the case that a platform a practice already owns may have the capability it needs and lack the person to run it. “You don’t have the bandwidth to actually monitor a certain stage of your process appropriately, and so if you changed, if you added headcount, or if you changed a process, the tool would work for you.” For early-stage practices her preference is fewer tools rather than more, because overlapping systems blur where work lives and mean paying for the same function twice. That is not an argument that one product should run a whole company. A practice needs accounting software and a learning management system alongside clinical operations and practice management, and no single tool covers all of it, so the buyer’s job is to decide where the business needs breadth and where it needs depth, then ask whether the systems chosen can talk to one another. Where they cannot, staff re-enter the same information and details fall into the gaps between tools.
At the summit she made a related argument about building software in-house, where the accounting is easy to underestimate. Ongoing maintenance, HIPAA compliance, and security attach to a system regardless of who wrote it, which is why she pushed back on the idea that cheap AI-assisted development has settled the build versus buy question. Buying carries its own exposure, as ABA providers were reminded when one automation vendor wound down and gave customers roughly 90 days to replace it.
How Office Puzzle Builds ABA Practice Management Software for Configuration, Not Customization
Office Puzzle began in 2018 as software written for a single Registered Behavior Technician who was overwhelmed by documentation, according to the company, and spread by word of mouth from there. Kurth described a first market that was narrow by accident and useful by design: the company built for Florida providers and their payers, and found that solving one customer’s payer-specific requirement often solved it for several others in the state. Because ABA requirements are regional, that pattern held until the customer base moved outside Florida, which is the tension the company is managing now.
The product decisions that follow are about configuration rather than customization. Rather than build separately for each customer, Kurth said, the company puts the complexity into setup. An administrator can set rules at the payer level, including whether overlapping appointments are permitted for a given insurer, so that day-to-day staff never have to know the rule. Scheduling populates documentation with what the system already knows, and flags a record that is not complete. Role-based views keep the configuration layer out of sight for staff who do not need it, so a technician doing data collection sees a graph, a plus and minus button, and a timer. In a field where a new hire is either productive in week one or lost in it, that is a retention argument as much as a design one, and at least one multi-site provider has found that cutting onboarding time nearly in half lowered early turnover instead of raising it. The company reports more than 800 practices and more than 30,000 ABA staff on the platform, at $19.99 per user per month, with a 30-day trial that includes one-on-one onboarding.
Kurth was careful about credit. “I don’t want to overstate my influence,” she said of a product philosophy that predates her arrival in March 2025. She described her own lane as customer experience and go-to-market, and her contribution as pushing the company toward self-service: trainings, materials, and onboarding straightforward enough that an owner working on a Saturday afternoon is not waiting for support hours. She frames that as a balance rather than a substitution. Owners increasingly expect to solve small things on their own schedule, and the company still answers when they cannot, so the problem she is working on is how to scale service as the customer base grows without diluting it.
The Pre-Mortem: A Change Management Habit for ABA Software Implementation
Kurth spent nearly six years at SimplePractice, the practice management platform for health and wellness providers, where she was Chief of Staff and then Vice President of Operations and Culture. Before that, at the real estate investment trust Prologis, she led a global rollout of a new data warehouse and business intelligence tools that required adoption across thousands of stakeholders in the United States, Europe, and Asia. The throughline, she said, has been change management, and she offered three habits to practices about to implement anything.
The first is ownership: somebody has to be responsible for getting the thing adopted, the lanes have to be defined if the work is shared, and the rest of the company has to know whose job it is. The second is restraint in design. Process-oriented people tend to launch with every feature turned on, and she argues for the opposite. “What is the core? I call it process lite,” she said. Get people comfortable with the foundation, document it well enough that it outlives its author, and add later. The third is planning by milestone. She has watched teams build hour-by-hour plans for a three-month project, then spend the rest of the quarter rebuilding the spreadsheet after the first week goes sideways, and she counts herself among them. “The planning is not the point,” she said.
Her favorite technique is the pre-mortem, which she found in Harvard Business Review and has used at every job since. The exercise, published by the cognitive psychologist Gary Klein in 2007, asks a team to assume a project has already failed and to explain why. Klein’s case for it rests on research by Deborah Mitchell, Jay Russo, and Nancy Pennington, who found that prospective hindsight, meaning imagining that an event has already occurred, improved the ability to identify reasons for an outcome by about 30 percent.
Kurth stages it. It is 2027, the practice management implementation has failed, clients are gone, claims were denied. What happened? “You almost are rewarded for thinking of how catastrophic something can go, in a contained environment,” she said. The output is a list of fears in a form a project owner can act on. One participant worries that technicians will never learn the tool while carrying a full schedule of clients, which becomes a training day with no sessions on the calendar. Not everything that surfaces is real, and the next step is triage by likelihood and impact. “People will catastrophize, but then they get it off their chest, and now you can see: is it a real problem? Is it an emotional problem?”
The audience for all of it, in her case, skews small. Roughly 20 to 25 percent of Office Puzzle’s customers run 100 or more staff, Kurth said, and most of the rest are early-stage or growing practices, often Board Certified Behavior Analysts who left to start something of their own. Those are the operators least likely to have a project manager on the payroll, and the most likely to end up as the one person quietly holding the system together.
