Every Goodbye Needs Somewhere To Go: The Felt Separation Phase & Why It May Be The Most Important Part of Therapy

She had been a therapist for eleven years when she described it to me. Not the session that broke her. Not the suicidal client or the disclosure that came at four-forty on a Friday. It was the client who just... stopped booking.

Twelve sessions. Good work by any measure. Then nothing. No ending. No goodbye. Just a gap in the schedule where a person used to be.

She said: "I still think about her sometimes. I don't even know why."

I know why.

No one teaches us about endings in graduate school, not really anyway. We learn termination as a clinical protocol: summarize gains, reinforce coping, provide referrals, close the file. What we don't learn is what to do with what gets left behind in us when a client walks out the door.

Not the dramatic doors. Not the ones with clean closure and a warm final session and a client who thanks you with a handwritten card. The other ones. The client who said "I think I'm good" after six sessions and you weren't sure you agreed. The transfer you didn't have time to process. The termination that was technically complete but emotionally unfinished. The session that ended, as they all do, three minutes before the next person arrived.

These endings accumulate. Over years. With nowhere to go.

This is what I want to talk about.

The Cycle Has Four Phases. We Only Train For Two.

The Cycle of Caring, developed by Skovholt (2005), describes the helping relationship as a repeated movement through four distinct phases: empathic attachment, active involvement, felt separation, and recreation. Once you see it, you cannot unsee it.

Empathic attachment is the opening, we meet someone, we orient toward them, we attune. It is other-directed by design. The relationship is organized around their growth, their healing, their life. Active involvement is the middle: the relational work, the intervention, the tracking and holding and confronting. This is what graduate programs train us for, what gets represented in textbooks and practicum supervision.

Recreation is the fourth phase, the return to self, the replenishment before the next cycle begins.

And then there is felt separation.

Felt separation is where the other-directed focus ends. It is the letting go. It is what happens (clinically, emotionally, sometimes physically) when that client walks out the door. Whether after one session or twenty. Whether after a clean goodbye or an ambiguous fade. It is the third phase of a cycle we repeat dozens of times a day, for decades of a career. And it is the phase we are least trained to inhabit.

What Accumulates When We Don't Attend to Endings

Rosenzweig, Farber, and Geller (1996) documented something meaningful: over the course of therapy, clients develop an internal representation of their therapist, an inner therapist they can carry between sessions, a part that can comfort and continue the dialogue long after the work ends. That's what we mean when we say we work ourselves out of a job.

But here is the question that research doesn't often ask: what do we develop? Who holds the representation of the client? And where does that attachment go when the work ends?

More often than not: nowhere organized. It dissipates, or it accumulates.

Figley (2002) framed the baseline riskas chronic inattention to self-care, combined with ongoing exposure to the suffering of others, creates conditions for profound emotional exhaustion. But I want to add something to that, because felt separation fatigue is not the same mechanism as secondary traumatic stress, even though they can co-occur. Secondary traumatic stress is about content exposure, absorbing the traumatic material we hear and witness.

Felt separation fatigue is about relational cost. It is the weight of attaching and letting go, over and over again, without adequate processing. It is not about what the client brought into the room. It is about what we are left with when they leave it.

When we skip the ending we don't return to a regulated baseline between sessions. We stay partially activated, partially attached. Physiologically carrying work into the next space where we were supposed to have recovered. Over years, that chronic low-level activation becomes a feeder into burnout. Not because we're doing it wrong. Because the structure of our work doesn't account for the full cycle of caring.

When the Ending Carries More Weight

For those of us who work with complex trauma, felt separation is not neutral territory.

Papa and colleagues (2024) described how even minor relational shifts such as a missed session or a slight change in availability, can activate intense fear of abandonment in clients with complex trauma histories. Their attachment systems read ordinary clinical distance as threat.

Hold that alongside what we know about our clients' histories with systems of care, not just family rupture, not just relational trauma, but abandonment by institutions, by helping professionals, by the very structures that were supposed to be safe. We are trying to be the corrective experience. The one who shows up. The one who stays.

While also being the one who will eventually leave. Who is supposed to leave.

Tanrıkulu and Gülüm (2025) found that termination processes can reactivate core attachment wounds triggering preemptive withdrawal, feelings of unworthiness, disrupted closure. So when we work toward an ending with someone who has known that kind of wound, we are doing something that is clinically demanding for them and emotionally demanding for us. Because we care. We've been in the room with everything they survived. And that ending asks us to trust that they can carry what we built together not just an inner therapist, but sometimes an inner parent, an inner nurturer, an inner protector.

That kind of ending deserves reflective space. It usually doesn't get it.

Werbart and colleagues (2019) found that therapists reflecting on work with clients who didn't improve were left with confusion, self-blame, and unresolved professional grief, particularly when the ending was premature or forced. The questions don't leave: Did I do enough? What went wrong? What if this happens again? And Piselli, Halgin, and MacEwan (2011) documented how therapists' own contributions to premature termination, the ruptures they didn't catch, the moments they misread, generate their own form of grief that rarely gets processed in consultation or supervision.

What Actually Helps

The research on felt separation points consistently in the same directions. None of them are about working less or thinking differently or building better walls.

Reflective processing of endings, specifically.

Thomas and Otis (2010) found that emotional separation, the regulated, boundaried, self-aware management of feelings that arise in clinical work, was associated with significantly less burnout and compassion fatigue and more compassion satisfaction. Not detachment. Not suppression. The active, skilled processing of what endings bring up. It is a learnable clinical skill, and like all clinical skills, it requires practice and space.

Intentional termination work beginning with the end in mind.

Structured endings are better for clients and for us. Not because tidiness is a virtue, but because Lavik and colleagues (2022) found that relational processes in psychotherapy, including how endings are handled, shape the therapeutic alliance in both directions: for the client and for the clinician's experience of the work.

When I begin a new therapeutic relationship, I'm asking early: What would it look like for you to be done? What do you want this to feel like at the end? We're creating a map together. For complex trauma clients especially, naming the ending from the beginning is part of the corrective relational experience. Chernus (2016) documented how even forced terminations, handled with intention, can be metabolized rather than simply endured by clients and by clinicians.

A structured reflective space.

Stacey and colleagues (2020) found that resilience-based clinical supervision was associated with reduced compassion fatigue, enhanced self-care, and clinician retention. People who had a place to process the weight of the work stayed in the profession. That finding should be in every training program.

Silverman and Segall (2024) found the same pattern in music therapists: career longevity was consistently linked to ongoing personal therapy and professional community.

What You Can Do Today

The felt separation phase is not only a passage for the client. It is a development site for the clinician. It is where the meaning of this work either consolidates or erodes. Where professional identity is either reinforced or quietly dismantled. Where the grief and pride and satisfaction of what we do get to be felt, if we have somewhere to feel them.

If we have that space, we can move into recreation: the return to self, the restoration of capacity, the possibility of beginning again. We cannot return to empathic attunement if we haven't moved through felt separation. Career longevity in the helping professions requires attending to self-care and burnout prevention across all phases of the cycle.

So here is what I want to offer:

• Start noticing how you end sessions. Not how your clients walk out. How you feel when they do. Is there a shift in your body? Heaviness, relief, something incomplete?

• Create a micro-ritual at the end of the day, or even between sessions. It doesn't need to be elaborate. A breath. A few minutes of silence before opening your notes. The walk between your office and your car. Something that signals: the active involvement phase is complete and the recreation phase has begun.

• Bring terminations into consultation, not just the hard cases, but the ordinary ones. The client who graduated after two years. The one who never booked again and you're still wondering. The one you're thinking about six months later. Those endings need somewhere to go. Consultation is that somewhere.

• And if you work with complex trauma clients: consider the ending from the beginning. Name it. Make room for it. Let it be part of the healing.

Your capacity to grieve your clients is not a liability or a sign that you're too close, too enmeshed, too soft for this work. It is evidence of the depth of the therapeutic relationship and of the humanness you brought into the room.

Every goodbye you've navigated has shaped who you are as a clinician. The question was never whether those endings mattered, the question is only whether you've had somewhere to take them.

If this resonated, share it with a colleague who might need language for what they're carrying. And if you want a place to land, I run a free monthly peer group called the Reset Circle, a grounded space for those of us doing this work. https://www.relitpractice.com/circle

References

Chernus, L. (2016). A self psychologist approaches retirement: "Forced" termination with highly vulnerable clients. Clinical Social Work Journal, 44, 150–159. https://doi.org/10.1007/s10615-015-0568-8

Figley, C. (2002). Compassion fatigue: Psychotherapists' chronic lack of self care. Journal of Clinical Psychology, 58(11), 1433–1441. https://doi.org/10.1002/jclp.10090

Lavik, K., McAleavey, A., Kvendseth, E., & Moltu, C. (2022). Relationship and alliance formation processes in psychotherapy: A dual-perspective qualitative study. Frontiers in Psychology, 13. https://doi.org/10.3389/fpsyg.2022.915932

Papa, C., Pugliese, E., Perdighe, C., Fimiani, R., & Mancini, F. (2024). "I am longing and afraid to depend on you": A case report on breakdowns of therapeutic alliance and interpersonal cycles in complex trauma. Brain Sciences, 14. https://doi.org/10.3390/brainsci14121207

Piselli, A., Halgin, R., & MacEwan, G. (2011). What went wrong? Therapists' reflections on their role in premature termination. Psychotherapy Research, 21, 400–415. https://doi.org/10.1080/10503307.2011.573819

Rosenzweig, D., Farber, B., & Geller, J. (1996). Clients' representations of their therapists over the course of psychotherapy. Journal of Clinical Psychology, 52(2), 197–207. https://doi.org/10.1002/(sici)1097-4679(199603)52:2<197::aid-jclp11>3.0.co;2-f

Silverman, M., & Segall, L. (2024). "I can't think of anything more fulfilling": An interpretative phenomenological analysis of eight female-identifying music therapists' career longevity. Journal of Music Therapy. https://doi.org/10.1093/jmt/thad028

Skovholt, T. (2005). The cycle of caring: A model of expertise in the helping professions. Journal of Mental Health Counseling, 27, 82–93. https://doi.org/10.17744/mehc.27.1.mj5rcvy6c713tafw

Skovholt, T., Grier, T., & Hanson, M. (2001). Career counseling for longevity: Self-care and burnout prevention strategies for counselor resilience. Journal of Career Development, 27, 167–176. https://doi.org/10.1177/089484530102700303

Stacey, A., et al. (2020). Resilience-based clinical supervision. Nurse Education in Practice. https://www.sciencedirect.com/science/article/pii/S0260691720314143

Tanrıkulu, M., & Gülüm, I. (2025). When therapy ends: A qualitative study on termination processes in youth with a history of residential care. Counselling and Psychotherapy Research. https://doi.org/10.1002/capr.70044

Thomas, J., & Otis, M. (2010). Intrapsychic correlates of professional quality of life. Traumatology. https://consensus.app/papers/intrapsychic-correlates-of-professional-quality-of-life-thomas-otis/fcd5d37529265a1c8fd442f7756f092b/

Werbart, A., Von Below, C., Engqvist, K., & Lind, S. (2019). "It was like having half of the patient in therapy": Therapists of nonimproved patients looking back on their work. Psychotherapy Research, 29, 894–907. https://doi.org/10.1080/10503307.2018.1453621