What to Look for in a Clinical Supervisor: Evidence-Based Guidance for CBT and ACT Therapists
- Christian Hughes
- 5 days ago
- 7 min read
Most therapists choose a supervisor the way they might choose a plumber. You check they're qualified, you check they work with your client group, and if they seem reasonable in the initial conversation you say yes. The decision is often not more deliberate than that.
This matters, because the quality of supervision has significant consequences for your development as a clinician, and not all supervision is equal. The evidence is clear that the right supervisor can deepen your clinical thinking, expand your repertoire, and fundamentally change how you relate to your work. The wrong one, or simply the inadequate one, can leave you technically ticking the accreditation box while your actual practice stagnates.
This post draws on a recently published Delphi study by Watson and colleagues (2026), published in the Journal of Contextual Behavioral Science, which sought to establish what expert CB supervisors actually agree on when it comes to good supervision practice. The findings offer a useful framework for any CBT or ACT therapist trying to make a more informed choice.
Why choosing well matters more than most therapists realise
Clinical supervision is often framed primarily as a safeguarding mechanism. It keeps practice safe, meets accreditation requirements, provides a space to flag risk. These functions matter. But framing supervision solely around safety creates a floor, but does little to speak to development. It tells you the minimum supervision needs to do, not what it could do to improve your practice.
Watson et al. (2026) describe the central purpose of contextual behavioral supervision as developing supervisee skills and knowledge in service of better client outcomes. Not simply keeping practice safe or meeting accreditation requirements (as much as those things matter). Developing you as a clinician, so that your clients benefit.
When you hold that as the standard, the question of who you choose as a supervisor becomes considerably more important.
What the research identifies as central to good CB supervision
The Watson et al. (2026) Delphi study brought together twenty expert practitioners who have published or presented on CB supervision, asking them to describe what distinguishes supervision from a contextual behavioral perspective and where they could reach consensus. Four themes emerged, each with specific implications for what to look for.
Functional analysis, not just technique review
The finding with perhaps the strongest consensus in the study was the centrality of functional analysis to CB supervision. Not functional analysis as a box on an assessment form, but as a genuine orienting principle in how cases are understood and how the supervisory conversation is structured.
In practice this means a supervisor who asks not just what is happening in your clinical work but why. What function does the client's behavior serve in context? What function does your behavior as a therapist serve? When an intervention isn't working, the question isn't primarily "what else could you try?" but "what does this tell us about our formulation of the problem?"
Participants in the study also described applying functional analysis to supervisory behavior itself, considering how the supervisor's own responses function in the supervisory relationship and whether they are producing the intended effects. A supervisor with this orientation is thinking about the impact of their feedback, not just delivering it.
The practical implication: when speaking to a potential supervisor, ask how they approach cases that aren't progressing. A supervisor who responds primarily with a list of alternative interventions is working at the surface. One who starts from the formulation is working at the level the research identifies as central.
Experiential learning over didactic instruction
Watson et al. (2026) found strong consensus that CB supervision should prioritise experiential over purely didactic methods. The phrase that emerged repeatedly from participants captures it well: "show, don't tell."
This doesn't mean supervision should never involve teaching. It means that developing clinical competence through direct experience, through roleplay, modelling, perspective-taking exercises, and working with the actual felt sense of the therapeutic interaction, is more effective than being told what to do and why.
A supervisor who primarily listens, reflects, and advises is delivering one kind of supervision. A supervisor who will roleplay a difficult clinical moment with you, model an approach, invite you to notice your own responses in the room, and use what emerges as data is delivering something considerably more active and more developmental.
This has implications for what to look for in an initial conversation. Does the potential supervisor describe their approach in terms of active, experiential work? Are they willing to use methods beyond discussion in sessions? Do they have a sense of what they would do with a stuck case beyond talking it through?
The supervisory relationship as a learning context in itself
The supervisory relationship was identified as the most significant contextual factor in the study, with 100% of participants agreeing that it constitutes a significant learning context. This isn't simply saying that a good relationship helps. It's saying that the relationship itself is part of how learning happens.
Watson et al. (2026) describe participants emphasising safety and equality as fundamental features, creating what one participant called 'a safe base for travelling together.' This included specific attention to the power differential inherent in supervision. Participants noted that power imbalance requires structured and deliberate attention, while the relationship should still feel genuinely collaborative rather than primarily hierarchical.
The concept of parallel process is relevant here, though not explicitly named in this paper. The relational dynamics that emerge in supervision, the way you bring your work, what you present and what you hold back, how you respond to challenge, often mirror dynamics in your clinical work. A supervisor who is attuned to this can use what emerges in the supervisory relationship itself as clinically useful data. One who isn't may inadvertently replicate unhelpful patterns without either party noticing.
When evaluating a potential supervisor, consider how the initial conversation feels. Does it feel collaborative? Is there genuine curiosity about your work and your development, or primarily an assessment of your competence? Is there a sense that both perspectives will be valued?
Values-driven practice, not rule-following
Watson et al. (2026) describe participants identifying values as guides to behavior across all aspects of CB supervision, from how the relationship is structured to how feedback is given. Explicit values clarification early in a supervisory relationship was seen as important, both in terms of the supervisor understanding your professional goals and values, and in terms of framing feedback within those values to reduce defensiveness and increase receptivity.
One participant put it plainly: framing feedback within a therapist's values, in service of their professional development and client outcomes, makes it easier to receive criticism without it feeling like an attack on competence or identity.
This has a practical implication. A supervisor who opens with questions about what you want from supervision, what kind of clinician you are trying to become, and what matters to you professionally, is setting up a values-informed supervisory relationship. One who moves straight to caseload review is not.
What strong consensus looks like in practice
The Delphi phase of Watson et al. (2026) produced eighteen statements that met the consensus threshold. Several are directly relevant to what to look for when choosing a supervisor:
The supervisory relationship should be experienced as a safe base. Supervision should be a collaborative process. A supervisor should have practical experience in using CB interventions. Functional analysis should be utilised in case conceptualisation. Regular and explicit feedback should be sought to ensure supervision remains effective and on task. Supervisors should embody willingness and flexibility to provide a social learning context.
These aren't aspirational ideals. They're the areas where experienced CB practitioners with publishing and presentation records in this field reached clear agreement. They can function as a checklist.
Questions worth asking before committing to a supervisory relationship
Drawing on both the Watson et al. (2026) findings and what the broader supervision literature tells us about what predicts supervisee development, the following questions are worth raising directly with a potential supervisor:
How do you approach cases that aren't progressing? This cuts to whether they work at the level of formulation and mechanism or primarily at the level of technique.
What does a typical supervision session look like for you? This surfaces whether the approach is primarily conversational or includes more active, experiential methods.
How do you think about the supervisory relationship itself? A supervisor who has considered this carefully will have something specific to say. One who hasn't will give a vague answer about being supportive.
How do you give feedback, and how do you think about the impact it has? This touches on the functional analysis of supervisory behavior that Watson et al. (2026) identify as important.
What do you want to know about me as a clinician at the start of our work together? This tells you whether they're interested in your development as a whole or primarily in your caseload.
The answers to these questions won't produce a perfect decision, but they will give you considerably more information than checking qualifications and reading a website.
A note on fit
Watson et al. (2026) describe CB supervision as attending to the personal histories and current experiences of both supervisor and supervisee, recognising that these shape how supervision unfolds. This is a reminder that supervisory fit is genuinely two-directional.
The right supervisor for someone earlier in their career, building foundational competencies, is not necessarily the right supervisor for someone mid-career working to deepen sophisticated clinical thinking. The right supervisor for someone primarily working with trauma presentations is not necessarily the right supervisor for someone working across a broad range of difficulties. And the right supervisor for someone who needs more challenge is not necessarily the right supervisor for someone who needs more containment.
Thinking about your own current developmental needs, rather than simply looking for the most credentialed supervisor available, is part of making a good choice.
Working with me
I offer clinical supervision for CBT and ACT therapists, with a focus on formulation-driven, functionally-grounded clinical thinking. My approach draws directly on the contextual behavioral science principles described in Watson et al. (2026): functional analysis of client, therapist, and supervisory behavior; experiential methods including roleplay and modelling; a collaborative, values-informed supervisory relationship; and explicit attention to your development as a clinician rather than primarily your caseload.
My background spans NHS specialist services, military mental health, and private practice, with particular expertise in trauma, complex presentations, moral injury, and ACT and contextual behavioural science. I'm a BABCP-accredited clinician and can provide supervision supporting therapists toward or maintaining BABCP accreditation.
If you'd like to discuss whether working together might be the right fit, a free 15-minute call is available.
