Defining the Applied Behavioral Scientist Role

A simple framework

Connor Joyce
Behavioral Design Hub
9 min readMay 3, 2022


Job postings for applied behavioral scientists have grown significantly over the past few years. This growth comes as companies begin to understand the value of behavioral scientists across departments. Scanning these job postings, one quickly realizes the field does not have a solid definition of an applied behavioral scientist.

I work as an applied behavioral scientist, where I help develop features to increase workers’ well-being and productivity. I have also collaborated with other behavioral and data scientists to establish my company’s first behavioral science team. These experiences have given me an understanding of how applied behavioral science can differentiate itself from more established groups (such as User Experience and Data Science) to bring unique value to an organization. And from this, I’ve developed the following framework to define the different roles of the applied behavioral scientist.

Three pillars of the applied behavioral scientist:

This simple framework provides a view of the main pillars to work towards as you establish yourself as an applied behavioral scientist. It can also be used as a template to develop a team's competencies to provide unique value. You need to establish skillsets within three pillars:

  • Behavioral Science domain knowledge
  • Mixed-method research experience
  • Communication and translation of research insights

Domain Knowledge

Becoming a behavioral scientist requires understanding the foundations of behavioral theory, causing the revolution we are all experiencing. There are three fundamental areas: Biases and Heuristics, Choice Architecture, and Social Norms.

Economics has been the social science explaining human decision-making for most of our current culture. Rooted in foundational tenets of a rational actor, many governments and business policies contain applied economic theories. The idea of behavioral economics upended this, suggesting there are fundamental biases and heuristics that humans use to reduce the cognitive load of modern life. The result is that we are not always rational. Studies conducted by Daniel Kahneman and Amos Tversky popularized this work. For more information, read Thinking Fast and Slow, and to learn the story of how it happened, check out the Undoing Project.

Further breaking from the traditional view of humans as rational decision-makers, Richard Thaler and Cass Sunstein, wrote the now uniformly popular book Nudge. Their work explored how a choice architect (an individual who controls the choice environment) can promote specific behaviors depending on how they set up the choice context. Minor tweaks to a choice environment that still allow freedom of choice but encourage one option over others are called nudges. It is essential to call out that nudging is just one intervention type of many in the Applied Behavioral Scientist tool kit, albeit a significantly valuable one. Nudge and many articles published simultaneously focused on simple tweaks with significant effects, such as changing organ donation defaults to opt-out. Newer research indicates nudges can be even more powerful but require personalization for the individual decision-maker. Review the latest edition of Nudge and this article on personalization for more information.

The latest tenant of behavioral science foundational research spawned the realization that behavioral change is social. Leading the field and becoming a popular press book is Robert Cialdini’s Influence. This book and the surrounding research suggest that many different factors influence why people ultimately make their decisions. The book’s insights, popular among salespeople, are beneficial for creating environments where social influence can encourage confident choices. Taking a more systematic approach by creating multiple types of social influence that make norms, Cristina Bicchieri is another leading player in the ever-evolving social influence science. For more information on her work, check out Norms In the Wild.

Knowledge examples from job postings:

Spotify — Behavioral Scientist:

  • You have a degree in behavioral economics, social or cognitive psychology, sociology, human-computer interaction, or a related field or gained your experience on the job.

Tonal — Behavioral Scientist:

  • Be the subject matter expert for behavioral science within Tonal & Provide scientifically informed and practically relevant behavioral science insights to multiple teams, including working closely with the Curriculum Intelligence team to inform product development.

World Resources Institute — Senior Behavioral Scientist

  • A deep understanding of the behavioral science literature and how it can be applied to help solve complex behavior change problems

Altria — Senior Behavioral Researcher:

  • Strong knowledge of consumer research, methodology strengths and limitations, and ability to strategically address business questions

Mixed-Methods Researcher

Communicating insights and knowledge of the field is essential to making insights actionable. One must also create insights in the first place. Generating research occurs through designing and running studies. Anyone in an applied setting quickly realizes that randomized controlled trials (the gold standard of research) are ideal but rarely feasible. Instead, the applied behavioral scientist must prepare such a study while recognizing tradeoffs that can speed up the process and decrease the required resources. Within the field of user experience, this has taken the moniker of the mixed-method researcher.

A mixed-method researcher can pull off both quantitative and qualitative studies. To be considered a solid mixed-method researcher, one should have competencies in at least one study from both sides but be familiar with many setups overall.

Qualitative Studies

The most common type of qualitative study an applied behavioral scientist will use is surveying. Surveys have many applications, from understanding the drivers causing a user’s action to measuring the impact created. The questions can take many forms, with some entirely open-ended, allowing users to run through their thoughts by utilizing Likert Scales to place opinions within a bounded scale. Closely following surveys in usage are interviews, which can take a similar form but allow for more follow-up questions and freedom with how the attendee responds. Specifically, interviews can take both a moderated and unmoderated approach. The former allows the interviewer to take the conversation in the best direction. The latter does not have an active interviewer, so the questions remain constant across the group.

Qualitative studies can go into further depth, such as with ethnographic studies and case studies; they can also break free from the individual, such as in artifact reviews. One example of an advanced qualitative method is a diary study which entails asking participants to take a survey or participate in an interview over a reoccurring period. The same questions get repeatedly requested to reduce the noise, and answers are given based on what the participant is experiencing in their environment. An applied behavioral scientist can use a diary study to drive after the thoughts, behaviors, habits, and concerns that participants are experiencing without directly asking them. Instead, they can use the temporal component of the study to see how a person expects to behave and then ask them how they did and why.

Quantitative Studies

Our world is rich with data; it is essential to create a model to understand what variables lead to an outcome and then utilize it to predict behavior. Just as crucial as surveys and interviews, applied behavioral scientists are highly likely to use regression equations throughout their work. Regression involves determining the relationship that one dependent variable has to one or multiple independent variables, with the most common type being linear regression. Building on the results of surveys, regression analysis can be used to understand better the effects of taking action on the user outcome data.

Regressions are not the only type of quantitative study; more advanced options include causal inference and machine learning models. One example of an advanced quantitative study is conjoint analysis. This method explores how people rank attributes of a topic against each other. It can take a multi-dimensional approach to understand what people value the most and what they are most likely to pay for or any different continuum. An applied behavioral scientist can use conjoint analysis to understand the variables within a problem set that people are most likely to want to change. Compare this analysis to variables people perceive as most changeable to assist in targeting the optimal starting point.

Mixed Methods examples from job postings:

Tonal — Behavioral Scientist

  • Conduct literature reviews, survey construction, interview design and delivery, and experimental design to inform evidence-based product and content development

Bayer — Applied Behavioral Scientist:

  • Demonstrated competency as a generalist in both qualitative and quantitative methods; you should be able to facilitate a user interview by yourself and run basic inferential statistics like correlations and simple T-Tests
  • Working knowledge of experimental methods emphasizes tradeoffs between certainty and resources like time and budget.

BetterUp — People Scientist:

  • Expertise in applied quantitative and qualitative methods (e.g., interview, focus groups, survey, RCTs, A/B testing, thematic analysis, ethnographic research).

Spotify — Behavioral Scientist:

  • You have solid quantitative research experience leading sophisticated behavioral research projects (e.g., questionnaires, sophisticated sampling, and weighing strategies).

Calm — Behavioral Research Scientist:

  • Ability to conduct data analysis and interpret findings.

GoodRx — Behavioral Science Analyst:

  • Lead research (quantitative and qualitative) to uncover user needs & motivations as part of a cross functional team consisting of user researchers, product managers, marketers, designers and engineers.

Communication Skills

The most intelligent person in the room is only influential if they can communicate their insights to a broader audience. This is especially true in an applied setting where competing priorities and stakeholders yield the most concise opinion capturing attention and funding.

Being a behavioral scientist in this environment requires tailoring insights to the audience. The same study can have multiple presentations depending on who will be the recipient. Take, for example, customer interviews around desired behaviors to change. It would be best to place study insights at the front when presenting to the requesting team. That same study should condense those insights into a simple slide or two for an executive audience and dive deep into the methodology and statistical methods taken for a group of researchers. Additional information such as setup and qualifying variables can be a part of the appendix.

Communicating insights from research is just one piece; serving as a translator of academic insights is just as important. This effort requires differentiating between valuable resources and marketing or thought pieces. A good applied behavioral scientist will incorporate all these sources, factoring in the quality of the insight from each resource without over tooling on one or two types.

A master’s degree does an excellent job of preparing an individual for this type of communication. It exposes one to academic-level research, building up their ability to understand nuance within peer-reviewed journal articles while equipping them with distilled insights into a digestible level for the general population. Research translation skills indeed can be achieved in other ways but finding the right program that exposes aspiring applied behavioral scientists to both sides will prepare them to be excellent communicators and translators.

Communication examples from job postings:

Calm — Behavioral Research Scientist:

  • Effective communication with scientific investigators and Calm teams about study design, development, implementation, results, and implications.

BetterUp — People Scientists:

  • Work in partnership with other researchers and functions within BetterUp (e.g., data analytics, design) to create comprehensive and coordinated research strategies that triangulate data from different sources/methods to develop recommendations.

Tonal — Behavioral Scientist:

  • Work effectively with the UX Research team to translate behavioral science theories and techniques into actionable UX research projects to inform product development

Intuit — Behavioral Scientist:

  • Excellent verbal and written communication and presentation skills

Once someone has achieved competencies in all three of these pillars, they will be ready to join a data science team, product management team, or many others as an applied behavioral scientist and quickly begin generating additional value for their team.

Please clap 👏👏 if you find this post helpful. Thanks!



Connor Joyce
Behavioral Design Hub

Mixed Methods Researcher and Behavioral Scientist. Ex-Microsoft, Twilio, Deloitte, and Tonal. On a mission to build products that change behavior! Penn MBDS '19