SCARF Model: David Rock's Framework for Leading People
Turn this article into takeaways for your work.
Each assistant summarizes the article only for you and suggests best practices for your work.
The SCARF model is a framework by David Rock that names five social domains the brain treats as a possible threat or reward: Status, Certainty, Autonomy, Relatedness, and Fairness. When a conversation or change pushes one of them down, people turn defensive and think less clearly. When it lifts one, they engage. Managers use it as a checklist before hard conversations.
Think about the last time you gave feedback and watched someone's face close. You probably didn't say anything false or unkind. But something in the exchange felt like a loss, maybe of standing, maybe of control, maybe of knowing what happens next. SCARF gives that feeling five names, which makes it something you can plan around.
What Is the SCARF Model?
Key Facts
- David Rock published SCARF in 2008 in the NeuroLeadership Journal, in a paper titled "SCARF: a brain-based model for collaborating with and influencing others".
- The paper rests on two themes it draws from social neuroscience: much social behavior is organized around minimizing threat and maximizing reward, and several social domains use the same brain networks as primary survival needs.
- Rock's own framing is that the model can be "applied (and tested)" in groups, so it is a synthesis of existing research, not a result from a trial of SCARF itself.
- A 2014 fMRI study of 60 participants challenged one of the premises behind the "social pain is physical pain" idea, finding distinct brain patterns for physical pain and for social rejection.
Rock was writing for people who lead, coach, and teach, not for neuroscientists. His claim is that the brain tags what it meets as something to approach or something to avoid, and that this tagging is fast and largely automatic. The five SCARF domains are his shortlist of the social situations that trigger it most.
In Rock's paper, the domains are defined like this: status is relative importance to others, certainty is being able to predict the future, autonomy is a sense of control over events, relatedness is a sense of safety with others (friend rather than foe), and fairness is a perception of fair exchanges between people.
The Five SCARF Domains
| Domain | The question the person is asking | Threat looks like | Reward looks like |
|---|---|---|---|
| Status | "Where do I stand compared with others?" | Public correction, unsolicited advice, being passed over | Recognition of improvement, being asked for expertise |
| Certainty | "Do I know what's coming?" | Vague expectations, rumors of a reorg, shifting priorities | Clear goals, a date when news will arrive, a defined next step |
| Autonomy | "Do I have any say in this?" | Micromanagement, a fixed plan handed down | Choice of method, ownership of a decision within limits |
| Relatedness | "Are these people on my side?" | Being excluded, working with strangers, a cold tone | A buddy, a mentor, small-group work, time to connect |
| Fairness | "Is this exchange even?" | Different rules for different people, hidden criteria | Transparent rules, a voice in how work is shared |
Status
Status is about relative importance, the pecking order. Rock notes that it can be threatened surprisingly easily, by giving advice or instructions, or by hinting that someone is slightly ineffective at a task. His practical suggestion for feedback is to let people assess their own performance first.
On the reward side, he argues that status doesn't have to come from promotion. People feel it rise when they see themselves learning and improving, and when someone notices. That's useful for managers with no promotion to give.
Certainty
The brain, in Rock's account, is constantly predicting the near future, and uncertainty pulls attention away from the task. Larger uncertainties, like not knowing your boss's expectations or whether your job is secure, can be highly debilitating.
You rarely can promise certainty. But you can shrink the unknown: say when the next update will come, break a project into steps, state the objective at the top of a meeting. Rock's own example for a restructure is to give a specific date when people will know more. Even if the news isn't settled, the waiting gets a boundary.
Autonomy
Autonomy is the sense of having choices and some control over your environment. Micromanagement is the textbook threat. Rock's smallest fix is almost trivial: offering "two options that could work, which would you prefer?" lands better than "here's what you have to do now."
He also stresses that autonomy works best inside agreed boundaries. Let people organize their own workflow or hours, but within parameters everyone knows. That's the same logic behind self-determination theory, where autonomy is one of three basic needs.
Relatedness
Relatedness is the quick judgment of whether someone is in your group or out of it. Rock argues that meeting someone unfamiliar tends to trigger a mild automatic threat response, and that trust and collaboration grow as safe connections do. His suggestions include buddy systems, mentoring, small action-learning groups, and, for distributed teams, shared personal stories or informal video time. At team scale, the Tribal Leadership stages describe how a group's shared language changes as those connections deepen.
This is where SCARF shades into psychological safety. A team where people feel like outsiders will also be a team where they don't speak up.
Fairness
Fairness is the perception that exchanges between people are even. Rock cites Tabibnia and Lieberman's work showing that fair exchanges are rewarding independent of the money involved. The underlying studies are worth reading directly: in Tabibnia, Satpute and Lieberman (2008), fair offers led to higher happiness ratings and reward-region activation than unfair offers of equal monetary value. A 2007 review by Tabibnia and Lieberman discusses what this means for organizations.
For managers, fairness mostly means transparency. People get angry at unexplained differences: one person gets the remote day, another doesn't, and nobody knows why. Publish the rules, and let teams help write them.
Threat vs Reward: Why It Matters for Managers
Rock's core argument is that threat and reward states change how well people think. He points to experiments suggesting that an approach state helps creative problem-solving and that a threat state narrows it. He also says the threat response is easier to set off than the reward response, which is why a manager's offhand comment can land so heavily.
Take that as a working hypothesis rather than a settled law. Here's what's well supported in plain terms, and what's more speculative:
| Claim | How firmly it's supported |
|---|---|
| People react badly to perceived unfairness, even at a cost to themselves | Backed by the fairness studies above |
| Social exclusion causes real distress, and brain imaging shows it | Eisenberger, Lieberman and Williams (2003) scanned participants excluded from a virtual ball game and found anterior cingulate activity tracked self-reported distress |
| Social pain and physical pain use the same brain systems | Contested: the 2014 study found separate patterns for pain and rejection |
| The five SCARF domains are the right and complete list of social triggers | Rock's synthesis; no controlled trial of SCARF itself is cited in his paper |
| Using SCARF improves team performance | Not established by the sources here; treat it as a planning aid |
So use SCARF as a way to ask better questions, not as a diagnosis machine. If someone goes quiet after a reorg announcement, SCARF suggests checking certainty and autonomy first. It doesn't tell you what's actually in their head. You still have to ask.
SCARF vs Other Frameworks
The model overlaps with several motivation and leadership frameworks. They answer different questions.
| Framework | Core idea | Level it works at | Use it when |
|---|---|---|---|
| SCARF | Five social domains trigger threat or reward in the moment | A single conversation or change | You're about to give feedback, announce a change, or run a tricky 1:1 |
| Psychological safety | Team members believe they won't be punished for speaking up | The team climate | You want people to raise problems and ideas |
| Self-determination theory | Autonomy, competence, and relatedness drive lasting motivation | The person's basic needs over time | You're designing roles or diagnosing disengagement |
| Maslow's hierarchy of needs | Needs are arranged in levels, from basic to self-actualization | The person's whole life situation | You're thinking about what a role must provide |
| Emotional intelligence in leadership | Leaders read and manage emotions in themselves and others | The leader's skills | You're building the self-awareness SCARF depends on |
A simple way to hold them together: psychological safety and self-determination describe the conditions you want over months. SCARF is the short checklist you run in the five minutes before a conversation.
How to Use SCARF in Feedback
Rock's own advice is that feedback often fails because it triggers a status threat, and that letting people assess their own performance first is one way to ease it. Here's a sequence that follows from that. The sequence is advice built on the model, not something Rock tested.
- Set the frame (certainty). Say what the conversation is about and how long it will take, so nobody spends the first five minutes guessing.
- Ask for their view first (status, autonomy). "How do you think the launch went?" gives them the first word and the chance to name problems themselves.
- Name specifics, not character (status). Describe what happened and its effect. "The report went out with last quarter's numbers" lands differently from "you're careless."
- Offer options (autonomy). "Two ways to catch this: a checklist or a second reader. Which fits you better?"
- Check it's even (fairness). If the standard applies to the whole team, say so. If it doesn't, expect them to notice.
- Close on the relationship (relatedness). End with what you'll do to help, and when you'll check in.
For a sharper take on candid feedback delivered with care, see radical candor.
How to Use SCARF in Change
Change touches certainty first, and usually autonomy second. A rollout plan that ignores both will meet resistance that looks irrational but isn't.
- Certainty: Share what is decided, what isn't, and the date you'll update again. "Silence until the plan is final" feels like a threat to many people.
- Autonomy: Give choices in how the change lands: pilot teams, timing windows, tool options. Leave the destination fixed and the route open.
- Status: Say who keeps their role and what expertise remains valued. People worry about being made obsolete before they worry about the new process.
- Relatedness: Keep teams together where you can, or pair people with a buddy through the transition.
- Fairness: Explain why this group changes now and that group later. Hidden criteria spread faster than the announcement.
How to Use SCARF in 1:1s
A 1:1 is a recurring chance to lower threat before it builds. A quick scan works.
| SCARF domain | Question to ask | Signal something's off |
|---|---|---|
| Status | "What work are you proudest of lately? What's felt invisible?" | Withdrawal, defensiveness, sudden perfectionism |
| Certainty | "What's unclear about priorities or your role right now?" | Repeated questions, delay, hedging |
| Autonomy | "Where do you want more say? Where do you want less?" | Compliance without ideas, passive resistance |
| Relatedness | "Who are you working well with? Who's hard to reach?" | Silence in meetings, working alone |
| Fairness | "Does the workload or recognition on the team feel even?" | Comparisons, resentment, side conversations |
You don't need all five every week. Rotate, or pick the domain that the last week's events most likely hit.
SCARF Prep Template
Copy this into your notes before a hard conversation or announcement.
| Domain | What could feel threatened? | What can I offer or protect? | Exact words I'll use |
|---|---|---|---|
| Status | |||
| Certainty | |||
| Autonomy | |||
| Relatedness | |||
| Fairness | |||
| Biggest risk domain | |||
| One follow-up and date |
Fill in the biggest-risk row first. If you can only address one domain well, choose the one most likely to be hit.
Examples by Business Function
These are illustrative scenarios, not case studies.
Sales. A manager reviewing a lost deal in front of the team threatens the rep's status. Move the review to a private 1:1, let the rep walk through what happened, and then share the lessons with the team without naming names. Publish the quota and territory rules so fairness complaints have nothing to feed on.
Operations. A new scheduling system arrives. Operators fear losing control over shifts (autonomy) and don't know how overtime will be assigned (certainty, fairness). Announce the rules for shift swaps and overtime before the launch date, and let each site choose its rollout week.
HR. A reorg is coming but details aren't final. Tell employees what has been decided, what hasn't, and the date of the next update. That one date does more for certainty than a vague reassurance.
Product and engineering. A product manager rewrites the roadmap without telling the team. Engineers lose certainty and autonomy at once. Share the reasoning, show what is fixed and what remains open, and let the team propose how to sequence the work.
Customer support. An agent gets a poor satisfaction score. Ask them to review the ticket first (status, autonomy), then agree one change together. Make sure the scoring rules apply equally to everyone on the team (fairness).
Do's and Don'ts
| Do | Don't |
|---|---|
| Scan the five domains before a hard conversation | Wing it and hope nobody feels judged |
| Ask people to assess their own work first | Open with "Can I give you some feedback?" and then pause |
| Give a date for the next update, even when news is thin | Go silent until everything is final |
| Offer two real options | Offer a fake choice with one acceptable answer |
| Explain the reasons behind differences in treatment | Let people guess why rules differ |
| Treat SCARF as a prompt for questions | Tell someone "you're just reacting to a status threat" |
| Praise specific improvement | Save recognition for promotion cycles |
That sixth row matters. Labeling someone else's reaction with a model, out loud, can itself feel like a status hit. Rock describes labeling as a technique for managing your own responses, so keep the vocabulary mostly in your own head.
Limitations to Keep in Mind
- It's a synthesis, not a tested intervention. Rock gathered findings from separate studies into a memorable frame. The sources here don't show controlled trials of managers using SCARF and getting better results.
- Some of the underlying science is debated. The claim that social and physical pain share a neural basis has been challenged, as in the 2014 study above.
- Five domains are a simplification. Real reactions mix them, and a missing one, such as plain workload, can matter more.
- Individuals differ. One person craves certainty, another prizes autonomy. Ask before assuming.
- It's easy to misuse. Reading it as a script for managing people rather than understanding them turns a helpful lens into a manipulation tool.
SCARF won't make hard conversations easy. What it does is slow you down for a minute before you speak, long enough to ask what the other person might stand to lose. Pick one upcoming conversation, run the five domains over it, and change one thing about how you open it.

Senior Operations & Growth Strategist
On this page
- What Is the SCARF Model?
- The Five SCARF Domains
- Status
- Certainty
- Autonomy
- Relatedness
- Fairness
- Threat vs Reward: Why It Matters for Managers
- SCARF vs Other Frameworks
- How to Use SCARF in Feedback
- How to Use SCARF in Change
- How to Use SCARF in 1:1s
- SCARF Prep Template
- Examples by Business Function
- Do's and Don'ts
- Limitations to Keep in Mind