Designing for Neurodiversity: What the Design World Can Learn from Applied Behaviour Analysis

Designers talk a great deal about user-centred design. They build personas. They conduct usability tests. They iterate based on data. They preach the gospel of understanding the user before building the solution. These are sound principles, and the design industry is better for having adopted them. But there is a field that has been practicing a version of user-centred design for decades, with a rigour and a depth of individualization that most design teams never approach, and it operates almost entirely outside the awareness of the design community. Applied behaviour analysis, the primary evidence-based therapy for autism spectrum disorder, is built on a methodology that mirrors the best practices of design thinking so closely that the parallels are striking once you see them. Observe the user. Collect data. Build an individualized solution. Test it. Measure the results. Iterate. Repeat. The difference is that in ABA, the stakes are not conversion rates or engagement metrics. They are a child’s ability to communicate, to navigate social situations, to develop independence, and to participate meaningfully in their own life.

For anyone in the design world who takes user-centred methodology seriously, the practices of applied behaviour analysis offer a fascinating and humbling case study in what it looks like to truly design for the individual. Not for a persona. Not for a segment. For one specific person, with one specific set of strengths, challenges, preferences, and processing patterns, observed and measured with a precision that would put most UX research departments to shame. Understanding how ABA practitioners approach the challenge of building individualized programs for children with autism illuminates principles that are directly transferable to design practice, and it reframes what “user-centred” can mean when the commitment to the user is absolute.

The Research Phase: Observation Over Assumption

Every ABA program begins with an assessment, and the depth of this assessment is something that designers would immediately recognize as analogous to their own research phase, only far more thorough. Before a single intervention is designed, the supervising clinician conducts extensive observation of the child across multiple settings and activities. They interview parents and caregivers to understand the child’s history, preferences, routines, and challenges. They use standardized assessment tools to establish baseline measurements of the child’s current skills across communication, social interaction, daily living, and behaviour. The result is not a generalized profile but a detailed, data-rich picture of this specific individual: what they can do, what they struggle with, what motivates them, what overwhelms them, and how they process information.

Compare this to how most design projects handle user research. A typical UX process might involve a handful of user interviews, a survey, and some analytics review. The output is a set of personas that represent aggregated patterns across user segments. These personas are useful, but they are abstractions. They describe a type of user, not an actual person. ABA takes the opposite approach. There are no personas. There is only this child, observed directly, measured precisely, and understood as an individual before any programming is designed. For designers who feel that their research phase could be deeper, more specific, and more directly connected to the solution they build, the ABA assessment model offers a benchmark for what thorough user research actually looks like.

Data-Driven Iteration: Measuring What Actually Works

If the assessment phase is the research, the ongoing data collection in ABA is the equivalent of continuous usability testing, except it happens during every single session, not as a periodic checkpoint. Every time a therapist works with a child, they record data on every skill being targeted: how many opportunities were presented, how many were completed successfully, what level of prompting was needed, and how the child’s performance compares to previous sessions. This data is not collected for reporting purposes. It is the mechanism by which the program is evaluated and refined in real time. If the data shows that a particular teaching approach is not producing progress, the supervising clinician modifies the approach. If the data shows that a goal has been mastered, the next skill in the developmental sequence is introduced.

This continuous feedback loop is what makes ABA a living, responsive system rather than a static curriculum. It is also what makes it so effective, because the program is constantly being optimized based on evidence rather than assumptions. Designers who practice A/B testing and iterative improvement will recognize the principle immediately, but the granularity of ABA data collection goes far beyond what most design teams attempt. BONDS Autism Centre, which has been providing ABA therapy and behaviour support services to families in the Burlington and Hamilton regions since 2016, collects session-level data on every child in their programs and reviews it regularly with families to ensure that the programming is producing measurable progress. This is not a quality assurance exercise. It is the core methodology. Every decision about what to teach, how to teach it, and when to change the approach is driven by what the data says, not by what anyone assumes or hopes is working.

Personalization at a Level Most Designers Never Attempt

The degree of individualization in ABA programming is something that the design world aspires to but rarely achieves. When designers talk about personalization, they usually mean adapting a standardized experience to user preferences: showing different content based on browsing history, adjusting a recommendation algorithm, or offering a choice between light and dark mode. These are useful features, but they are variations on a template. The underlying system is the same for everyone. ABA operates at a fundamentally different level of personalization. Every child’s program is built from scratch based on their individual assessment. The goals are specific to that child. The teaching strategies are selected based on how that child learns most effectively. The reinforcement systems are designed around what that child finds motivating. Even the pacing and structure of sessions are calibrated to that child’s attention span, sensory profile, and emotional regulation capacity.

This level of individualization is possible because ABA practitioners invest the time to understand each child as a complete individual before designing anything. They do not start with a standard program and make adjustments. They start with the child and build the program around them. For designers who work in fields where genuine personalization matters, whether that is healthcare technology, educational software, accessibility design, or any domain where one-size-fits-all solutions fail the people who need them most, the ABA model demonstrates what it takes to make individualization real rather than cosmetic. It requires deeper research, more granular data, more frequent iteration, and a willingness to build custom solutions rather than configuring standard ones.

Designing Across Environments

One of the most interesting parallels between ABA and design practice is the challenge of designing for multiple contexts. Designers building responsive websites must ensure that the experience works across desktops, tablets, and phones, each with different constraints and interaction patterns. ABA practitioners face an analogous challenge: ensuring that skills learned in one environment transfer to others. A child who learns to request a snack using words during a therapy session needs to be able to do the same thing at home, at school, and in the community. This generalization across environments does not happen automatically. It requires deliberate planning, practice across settings, and collaboration between the therapy team, parents, and other caregivers. Centres like the team at BONDS offer programming across centre-based, home-based, and community-based settings specifically because they understand that skills practised in only one environment tend to stay in that environment. The goal is not just learning but functional, transferable competence that works in the real world.

This is a principle that designers would benefit from internalizing more deeply. A solution that works perfectly in controlled conditions but fails in real-world use has not actually solved the problem. ABA practitioners deal with this challenge every day, designing interventions that must survive the noise, variability, and unpredictability of real life. The therapy session is the prototype. Home, school, and community are production. The gap between those two contexts is where most of the hard design work happens, and it is work that requires ongoing collaboration with the people who inhabit those environments rather than a handoff and a hope that things will translate.

Scaling from Individual to Group

ABA practice also offers a useful perspective on the tension between individual and group experiences. Most ABA therapy is delivered one-on-one, with a therapist working directly with a single child. But many centres also offer group programming where children work on social skills alongside peers of similar age and abilities. The design challenge in group programming is maintaining the individualization that makes ABA effective while creating an environment where children can practice the inherently social skills that one-on-one sessions cannot fully address. Each child in the group has their own goals, their own data tracking, and their own individualized approach, but the group setting provides the social context that those goals require. Taking turns, responding to peers, managing the unpredictability of group dynamics, these are skills that cannot be practised in isolation.

Designers who work on collaborative tools, multiplayer experiences, or any product where individual user needs must coexist with group functionality will recognize this tension. The best group experiences do not flatten individual differences. They create a structure flexible enough to accommodate them. ABA group programming achieves this by maintaining individualized goals within a shared environment, which is essentially what good collaborative design does: creating a framework that serves each participant’s needs while enabling the group dynamic that makes the experience valuable for everyone.

What Design Can Take from This

The design world does not need to adopt ABA methodology wholesale. The contexts are different, the users are different, and the constraints are different. But the principles that make ABA effective are the same principles that make design effective, taken to an extraordinary degree of commitment. Observe before you build. Measure what actually happens, not what you hope happens. Individualize relentlessly. Test across real-world environments, not just controlled ones. Iterate based on evidence. And above all, serve the person in front of you rather than the abstracted version of them that exists in your planning documents.

For designers who want to expand their understanding of what user-centred practice looks like at its most rigorous, the field of applied behaviour analysis is worth studying. Not because designers should become therapists, but because the methodological discipline that ABA demands, the insistence on data over intuition, on individual understanding over generalized assumptions, on real-world validation over lab-condition success, is exactly the discipline that separates good design from great design. The best ABA practitioners are, in a very real sense, designers. They observe, they build, they test, they iterate, and they measure their success not by how elegant the solution looks but by whether it actually works for the person it was built for. That is a standard worth aspiring to in any field where the work is ultimately about serving human needs.