What Solution Focused Therapy Actually Looks Like in Practice
Most people come into SFT expecting a lot of history-gathering, and that is the first thing you need to unlearn. The model assumes the past is mostly background noise unless it directly contains examples of when the problem was already solved. I spent years doing standard intake assessments before I encountered a client who came in for chronic anxiety with a detailed trauma timeline stretching back to childhood. We spent the entire first session redirecting her away from the narrative. She was frustrated. So was I, honestly. But within three sessions she identified that she had gone through two winters previously without panic attacks occurring. That was the data point we worked from. The goal became amplifying what was already working, not excavating why it broke. SFT is goal-driven by design, but the goals it targets are specific. They are not vague aspirations like "feeling better" or "being happier." The goals must be concrete, observable, and within the client's sphere of influence. When I run a session, I am looking for three things: a desired future the client can describe in behavioral terms, exceptions to the problem that already exist in their life, and a scaling metric they can use to track progress without me having to prescribe one. The miracle question is the standard entry point, and it sounds theatrical but it works mechanically. You ask the client to imagine waking up tomorrow and the problem is solved. The goal here is not fantasy fulfillment. It is extracting a picture of the preferred future that is specific enough to act on. "My anxiety would look different" is not useful. "I would be able to sleep through the night without checking the lock four times" is.
Then there is the scaling question. Usually on a one to ten scale, where one is the worst the problem has been and ten is the preferred future. This turns vague emotional states into trackable data. A client who says they are at a three and wants to get to a five gives you a clear target. You then ask what would move them from three to three point five. That is where the work actually happens. The goals are incremental. You are building momentum, not jumping to a cure. Exception-finding is the third pillar. Problems are not constant. There are moments when the issue is less severe or absent entirely. The goal here is to help the client notice those moments and then figure out what they were doing differently. If someone reports their depression lifts every Tuesday afternoon, you do not need to investigate deep childhood dynamics. You need to ask what is different about Tuesday afternoons and get them to repeat that behavior more often.
The Technical Mechanics Behind the Method
Under the hood, SFT draws heavily from social constructionism and the work of Steve de Shazer and Insoo Kim Berg at the Brief Family Therapy Center in Milwaukee. The model emerged in the late 1970s as a reaction against the dominant psychodynamic and behavioral paradigms, which both treated therapy as a lengthy excavation process. De Shazer noticed that his most effective clients were the ones who stopped talking about what was wrong and started describing what they wanted instead. He built an entire methodology around that observation. The terminology you will encounter includes the formula of the session, which is a structured way of framing the conversation from minute one. You start by asking what the client hopes to get out of the session. Then you move to the scaling question, then exception-finding, then the miracle question, and finally a feedback loop where you summarize what you heard and assign a small task. The typical structure takes about twenty minutes of direct questioning and leaves the rest to the client's own implementation. Coping questions are another tool worth mentioning. These are used when a client seems stuck or hopeless. Instead of asking what they want, you ask how they have managed to keep going despite the difficulty. This shifts the frame from deficit to survival and often reveals internal resources the client had stopped acknowledging. It is a subtle reframe that can unlock a session quickly.
Get the Full Details
A Problem I Ran Into That Most Guides Skip Over
I had a client once presenting with OCD symptoms centered on symmetry and ordering. Standard SFT exception-finding should have worked fine here, but it did not. The problem was that her exceptions were not meaningful to her. She would have a perfectly ordered desk for three days, then feel nothing because the anxiety never actually left. The goal became unattainable using traditional SFT language because her definition of "solved" was perfection, which is logically impossible. The workaround was to redefine the goal away from the absence of anxiety and toward tolerance. I asked her to describe what a six out of ten on the distress scale would look like behaviorally. She had to sit with a slightly messy desk and rate her anxiety afterward. That was the shift. The goal was no longer elimination. It was desensitization. We used SFT mechanics but blended in exposure principles without explicitly framing it that way. Two clients later I realized this was not unique to her. SFT struggles with conditions where the problem serves a rigid function and the client defines success in all-or-nothing terms.
Pitfalls and Where This Approach Fails Completely
SFT is not a universal tool. It breaks down in several scenarios that practitioners rarely admit publicly. It does not work well with acute psychosis, active substance dependence, or severe personality disorders where reality-testing itself is compromised. The model assumes a baseline of cognitive flexibility and future-orientation. If the client cannot genuinely imagine a different future, the miracle question becomes meaningless or even alienating. Another failure mode is with clients who expect directive advice. SFT therapists do not give advice. We ask questions. Some clients interpret this as incompetence or avoidance. I had a client leave after session two because she wanted me to tell her what to do about her marriage. I could not offer that. The model simply does not support it. In those cases, referring to a more directive modality like CBT or couples counseling is the honest move. There is also a risk of prematurely closing down legitimate grief or processing. If someone is dealing with actual loss and you start redirecting them toward solution-building, you are pathologizing normal emotional response. SFT can feel dismissive if applied without clinical judgment about what the client actually needs in the moment.
How to Actually Use This in a Session
Start the session by asking what has changed since the last meeting, or what they want to focus on today. Listen for any sign of progress, however small. Then pick the scaling question and anchor it to a specific domain. Relationship? Work? Sleep? Mood? Pick one. Do not try to scale everything at once. Move into exception-finding with a question like, "Tell me about a time recently when the problem was less intense. What was different?" Then drill down. What were they doing? Who was present? What was their mindset? The details matter. Generic exceptions are useless. Specific exceptions are replicable. When you get to the miracle question, keep it short and direct. Do not embellish. The longer the setup, the more the client will resist. Ask it, pause, and let them answer. If they struggle, offer a gentle prompt about their daily routine. What would they notice first if the problem disappeared?
Closing the session requires a summary and a small task. The task should be something the client can actually do before the next meeting. Not a breakthrough. Not a transformation. Something tiny. "Notice when the problem does not show up this week" is a valid task. "Fix your entire life" is not. The duration of treatment matters too. SFT is designed to be brief. Eight to twelve sessions is the standard range, though many clients show measurable change within three or four. If you are still doing lengthy exploratory work past session six, you have likely drifted into a different model without realizing it.
What the Research Actually Says
The evidence base is solid but not overwhelming. Meta-analyses published between 2015 and 2023 consistently show moderate effect sizes for anxiety and depression, with stronger outcomes for brief crisis interventions than for long-term chronic conditions. The model's strength is speed. It reduces session count by roughly forty percent compared to traditional approaches for uncomplicated cases. That is a real saving in cost and time, but it does not mean it is better for complex presentations. It means it is faster for straightforward ones. One thing the research does not emphasize enough is therapist fidelity. SFT looks simple on paper. It is not simple to execute well. The difference between a competent SFT session and a mediocre one often comes down to how precisely the therapist reframes and whether they resist the urge to drift into problem-talk. It requires discipline. So if you are considering this approach, test it on the right cases. Clients who are motivated, somewhat future-oriented, and dealing with adjustment issues or mild-to-moderate anxiety and depression will respond fastest. You will see results within weeks. For the rest, pick a different tool or combine modalities. Nothing about this model is wrong. It is just narrower than its proponents sometimes imply.
