Adherence, adoption, and engagement are behavior problems before they are product problems. We surface the behavioral barriers your dashboards record but never explain — and identify which interventions are most likely to work before expensive implementation begins.
Clinical outcomes aren't just about the molecule or the algorithm. They're about the human systems that adopt them. We bridge the gap between "it works in theory" and "it works in the workflow."
CFIR, RE-AIM, and NPT are good at what they were built for — mapping determinants, evaluating reach and maintenance, explaining how a practice becomes routine. We use them. But they were built to study organizations adopting practices, and the failures we get called in on happen one person at a time, before adoption ever begins.
Implementation science asks why a practice fails to take hold. We ask what the person was trying to accomplish before anyone offered them a practice — and that answer usually explains the first question.
Successful adoption isn't an accident. It's the result of systematically de-risking the gap between clinical intent and real-world application.
Intervention, practice, workflow, innovation
Individuals, teams, organizations, systems
Context, environment, constraints
Processes, strategies, sequencing, workarounds
Behavioral mechanisms, barriers, facilitators
Sustainability under real-world conditions
No single method answers an implementation question on its own. Synthetic research produces the first map — which barriers look like they bind, which strategies are worth evaluating, and what to actually ask — fast, and before you have spent anything on recruitment. Its unit is the archetype — constructed positions deliberately spread across a population, so the behavioral range is covered before you know which part of it matters. Human-participant research then works one person at a time: AI-moderated interviews for reach across individuals, 1:1 interviews for depth on any one of them. The expertise isn't in owning three methods. It's in knowing which uncertainty you are actually trying to remove — and at which level.
Formative by design, and archetype-based — not generic AI personas. Each archetype is a deliberate position in the population, spread across six-force demand psychology, stage of change, and personality range, then read with the same apparatus we bring to human data.
Arrive at the human research already knowing where to look. Formative only — every barrier and strategy it surfaces is a candidate for validation, never a finding.
Real individuals, conversationally interviewed at a scale and speed a 1:1 schedule cannot reach. The instrument is the one the synthetic phase helped shape, so each conversation starts further along.
Reach more individuals without widening the timeline. The same people you would interview yourself — more of them, sooner.
The deepest read on a single individual. Switching interviews run by hand, where the causal story and the moments that actually moved one person come out.
The deepest read on mechanism. Irreplaceable, and the reason the other two exist — they get you here faster and better prepared.
No — and we would not sell it to you on that basis. Synthetic work is formative, not confirmatory. It is fast and cheap enough to run before you have committed to a research design, which makes it excellent at mapping candidate barriers, surfacing strategies worth evaluating, and shaping the instrument. What it cannot do is tell you any of that is true of real people. That is what the human-participant modalities are for, and every study we run is scoped so the synthetic phase hands off to them.
Most research stops at findings. We run every study through three integrated disciplines—so evidence moves from demand, to diagnosis, to design.
The lens that reaches furthest upstream. Why a person would change at all, what progress they're hiring an intervention for, and the six forces — Pushes, Pulls, Desires, Avoidances, Habits, Anxieties — acting on that decision before any intervention exists.
Where adoption breaks, and at which stage. Every interview read through the stages of change, then against capability, opportunity, and motivation and the belief structures that govern health decisions.
Diagnosis translated into intervention strategy—making the target behavior easier, timelier, more attractive, and sustainable under real-world conditions.
Behavior is always a function of the Person and the Environment—change the environment before trying to change the person.
Two decision systems drive behavior—System 1 (fast, automatic) and System 2 (slow, deliberate). Most adoption runs on System 1.
Every behavior has an Activator and a Consequence—and consequences determine whether the behavior continues or stops.
Behavior follows the status quo unless Motivation, Ability, and a timely Prompt align at the same moment.
Behavior is shaped by Intrinsic motivation from inherently rewarding activity and Extrinsic motivation from rewards and avoidances.
Why it matters: single-framework research finds what its framework looks for. Integrating demand-side, diagnostic, and design-side disciplines is how we catch the failure modes each lens misses alone.
Your product works for the people who use it. The problem is how few of them keep using it — and your instrumentation records the drop-off without ever explaining it.
Find out why engagement decays →You've sold the engagement and the behavioral question sits inside it. Rather than building the research capability, you run it through us and deliver it as yours.
Talk through a live engagement →We work behind your brand. You own the client relationship and the deliverable carries your name. We don't approach your clients, and we don't compete for the engagement.
The program is sound and the population it's for is the hardest to reach. Uptake is decided by cost, distance, routine, and pride long before it's decided by the program design.
Pressure-test a program before rollout →Formative, implementation-focused research designed to inform better decisions before rollout and scale. We translate evidence into operational reality.
Structured for speed. Grounded in science.
We frame the behavior before we study it—defining the job, the population, and where adoption is most likely to break.
We deploy the modality mix the question demands—triangulating signal rather than relying on a single source.
We turn evidence into a decision—diagnosing the dominant forces and specifying the interventions that address them.
Typical engagement: 3–4 weeks · Scoped per intervention · Modality mix set in Phase 01
Every study lands as a set of things you can act on. The Living with COPD study is the worked example — each artifact below appears in it.
Which barrier binds at which stage, so you know where in the journey a person actually falls off — not a flat list of obstacles.
The functional, emotional, and social progress people are reaching for, stated in their terms rather than yours.
Pushes, Pulls, Desires, Avoidances, Habits, and Anxieties mapped across the cohort — what's driving toward change and what's holding against it.
Distinct participant profiles with personality range, showing where the population genuinely differs rather than averaging it away.
Each barrier sorted by kind, because a friction barrier and a belief barrier call for entirely different responses.
A named behavioral strategy against each binding barrier, so the finding arrives as something to build against.
The synthesis, the confidence you should place in it, and what it means for the decision you brought us.
What this is not. We identify which interventions are most likely to work before expensive implementation begins. We don't build the intervention, run the pilot, or own the rollout — those stay with your team, and we scope the research so it hands off cleanly to them.
Formative studies that map where behavior breaks down — and what would have to change for it to hold.
Roughly half of patients abandon a device they know works—the intervention succeeds clinically and fails behaviorally.
Roughly half of patients discontinue within the first year—and the barriers are behavioral, not clinical.
A condition that skews rural and low-income, where daily self-management collapses under routine disruption and constrained access.
For Digital Health, HealthTech, and Public Health teams where adherence and behavior change decide the outcome. Bring one behavior that isn't happening.
Book a discovery call →Serving Digital Health, HealthTech, and Public Health.
Seattle, WA • Remote
