The Solution Frame: 
The Genius of Solution-Focused Therapy

© 1999 John E. Perkins, III
5201 22nd Ave. NE Ste. 201
Seattle WA 98105
206 524.4496
Send email to the address on our home page:


Solution-Focused Therapy (SFT) takes a major departure from the problem-focused orientation in psychotherapy begun by Freud and perpetuated by most fields of therapy. The deepest roots of SFT can be traced back to Gregory Bateson and Milton Erickson. SFT seized as a founding truth an observation that Erickson made that the "sine qua non of psychotherapy should be the present and future adjustment of the patient" (Berg and Miller, 1992:13). Most therapies see clients as coming into therapy with problems seeking help; solution-focused therapists see clients as coming in with solutions seeking expression. This reversal of the usual paradigm is what interests me in SFT.

Insoo Kim Berg, along with her husband Steve de Shazer, founded the Brief Family Therapy Center in Milwaukee, Wisconsin. Berg and de Shazer have focused their attention on the types of questions therapists can use as tools to help clients get motivated to achieve their goals for therapy. I heard Berg present at a the Special Problems in Marital and Couples Therapy: Dealing Effectively with Difficult Couples Conference in 1992.

How This Paper is Organized

This paper is based on the ninety minute presentation I made to the Homeless Families Network on SFT on September 14, 1994. I am following the general structure of that presentation. After I present the basic philosophy of SFT and criteria for well-formed goals, I will discuss the concept of the miracle question. In the discussion of the miracle question, I will include a transcript of my demonstration of it at this workshop. After that, I will examine the other four useful questions which make up SFT. The last section of this paper takes a look at some of the feedback I have received both during and after the training.

Basic Philosophy

Therapists literally must learn to see with solution-seeing eyes and hear with solution-hearing ears and feel with solution-feeling emotions. They must learn how to trust that solution seeking part of their clients and help the client find ways of letting that part have a bigger say in their day-to-day lives. Four principles undergird the work of the solution-focused therapist:
A. Meet the client at his or her model of the world.

B. Transform the client from being a "visitor" or "complainant" to being a "customer."

C. Start with the end in mind (miracle question and well-formed outcome conditions).

D. If it ain't broke (in the client's mind) don't fix it.

Now to look at the principles in more detail. Principle A: Meeting the client at his or her model of the world. The way the client makes sense of her life is very important to her. What she does helps her life hold together, and make sense. If the therapist meets her at that level, then the client can trust the therapist. Then some of the resistance is avoided, and a lot of the repetitive returning to "old issues" is avoided. For example, a therapist might accept any first goal a client wishes to work towards, however "out of order" it might seem to the therapist. From this principle, a homeless drug-abusing pregnant client would be helped to solve her first priority first: getting housing, and not the therapist's or society's goal of getting her off drugs. The therapist must trust that in time, as successive goals are accomplished, the client will turn her attention to reducing her dependency on drugs.

Meeting the client at her model of the world naturally helps with Principle B: Transform the client from being a "visitor" or "complainant" into a "customer." A "customer" for therapy is ready to do the "work" of therapy because he or she recognizes, in his or her own way, the benefits of actively participating in therapy. A "visitor," at bottom, maintains a wariness or distance from the expectation that he or she might change. Sometimes when a client is sent by someone against his will, he arrives in a "visitor" mode, that is, he just physically shows up for his sessions without the intention of making any personal changes. A "complainant" comes into therapy complaining that someone else is responsible for their difficulties, or feels frustrated by their difficulties yet isn't ready to change. The client may feel no need to change. Visitors and complainants often are in therapy at the suggestion of someone and are less there out of their own choice or sense of needing help (Berg and Miller, 1992:22-29). This paper will focus on working with clients who are clearly "customers."

Principle C, start with the end in mind, gives most of the therapy schools derivative of Bateson their power. It is built into the name of this approach: Solution-Focused Therapy. In my brief training I emphasized that the goal must be one that the client wants, despite whatever the therapist might believe needs attention. In Berg and Miller's terminology this is called the hidden customer (1992:239-31). One of the litmus tests for whether clients are in the role of visitor, complainant or customer is how well they can participate in framing a destination for their therapy. Customers know what they want and accept working to get it. But, a visitor or complainant might be a hidden customer. Hidden customers have goals that they want, but for which they might not have been referred by their well-wishing friend, boss, principal, lover, spouse, or relative. The solution-focused therapist will follow the clients' desires and help them frame a goal meeting their preferences.

Principle D, if it ain't broke (in the client's mind) don't fix it, is a reminder to the therapist to respect the rights and wishes of the client. How clients see their lives, and the order in which they approach solutions must make sense to them. Respecting a client's wishes also neatly avoids a lot of "resistance" or "opposition offered by the patient to the orders, actions, recommendations, or suggestions of the therapist" (Harper, 1959:169).

Well-Formed Goals

SFT and other brief, client-based goal-oriented therapies take a lot of time developing and training practitioners in what makes a goal an achievable one and worthy of therapy. SFT has a set of guidelines for the therapist to measure the worthiness of a client-based therapeutic goal. A well-formed goal has seven qualities, according to Berg and Miller (1992, 32-44; de Jong and Miller, 1995:730-731).
1. The goal must be important to the client. This helps build on the cooperative atmosphere needed for successful therapy, respects the client, and brings the therapist into a dialogue with the client.

2. Keep goals small and achievable. This helps both client and therapist recognize progress. Within a therapy session, clients often begin with very large, vague goals, and one way the therapist can begin to help them think small is to ask what would be "the first small signs of success" on their way to the larger goal (Berg and Miller, 1992:36). Skilled therapists help clients name and accomplish a series of small goals which contribute to a bigger whole, since it is "easier to 'fill out a job application' than to 'get a job'" (de Jong and Miller, 1995:730).

3. Make goals concrete, specific and behavioral. John Grinder, one of the founders of Neuro-Linguistic Programming (NLP), often quipped in his training programs, "Would you recognize it if it tap danced on your foot?" In our ordinary everyday world, we use the qualities of specific and concrete many times a day, as when we say "I'm looking for my car keys, not my house keys," because just any key won't do the job. Berg and Miller say that non-behavioral goals, such as having more self-esteem, living a sober life-style, and getting in touch with feelings are difficult to achieve, mainly because success and progress are difficult to gauge (1992:37). A concrete goal, such as "praising my child twice this week when he makes his bed," can be counted, which makes determining success towards its accomplishment possible.

4. Goals express the presence of something or of a behavior, rather than an absence. "Goals must be stated in positive, proactive language about what the client will do instead of about what she will not do" (Berg and Miller, 1992:38, emphasis in original). Berg and Miller offer three reasons for defining goals in positive terms (:38-39). First, it helps clients and therapists determine when a goal has been met. Second, we are always doing something; if one behavior is absent something must be there to take its place. If a client will not be smoking, what will she be doing instead?

Third, sometimes commanding ourselves not to think of something builds an obsessive cycle where we conjure an image of the forbidden activity and then try to ignore it. The book White Bears and Other Unwanted Thoughts: Suppression, Obsession, and the Psychology of Mental Control (Wegner, 1989) is devoted to this particular trap we unwittingly spring on ourselves.

5. Goals are expressed as beginnings rather than endings. This helps clients and therapists find ways to be on "track" immediately (Walter and Peller, 1992:55). This supports a view of living and therapy as both being processes. The beginning and progress towards a goal can be better described and used for feedback than the ending or ultimate goal, which might be very far off in terms of time and effort. Naturally arising from this quality of goals are questions during therapy that support small improvements and beginnings. Examples of these type of questions are: "What would be the first small signs of change?" or "What small step could you take over the next week which could begin to move you towards your goal?"

6. The goals are realistic and achievable within the context of the client's life. This quality of well-formed goals reminds the therapist that a goal needs to nest well in the larger story the client calls his or her "life."

7. The client sees the goal as involving "hard work." This quality helps the client build a positive "face" and protects and promotes the client's feelings of dignity and self-worth (Berg and Miller, 1992:43). This allows the client to internalize personal responsibility for achieving the goal while having a self-respecting place to fall back to in case of failure.

Constructing goals as involving "hard work" parallels the client's perception of how difficult it has been previously to change. A goal needing "hard work" leaves the client and therapist in a win-win situation. If they fail to reach the goal, it is still possible to achieve it if they can find how to work a little harder. If the client quickly gets to the goal, she can be complimented on being able to figure out such a difficult solution in a short period of time. Slow, steady progress can be accepted as normal and the client can be praised for her hard work (Berg and Miller:42).

The Miracle Context

The miracle question anchors the way solution-focused therapists conduct all interviews. It frames each session, and the entire process as one of constructing new behaviors for the near to indefinite future. "This specially designed interview process orients the client away from the past and the problem and toward the future and a solution" (Berg and Miller, 1992:13). Actually, the phrase "miracle context" better captures the central importance of this concept to the SFT approach than "miracle question" does.

What is the miracle question? After some basic information gathering, the therapist asks

Suppose that one night, while you are asleep, there is a miracle and the problem that brought you here is solved. However, because you are asleep you don't know that the miracle has already happened. When you wake up in the morning, what will be different that will tell you that the miracle has taken place? What else? (Berg and Miller, 1992:13)
This neatly turns the client's attentions away from that old, unsuccessful, past behavior towards a future where the problem no longer exists. It can easily be slipped into the therapeutic conversation as this example, from Steve de Shazer working with a client who wanted to stop drinking, shows:
Client: I just caught myself just going to stay home and just quit cold turkey, but I ended up the same.

SdeS: Mm, hm.

C: I guess I can break the spell.

SdeS: Mm, hm. You can. So, I have a somewhat strange question, but, uh, suppose that ah...when you go home tonight and you go to bed and you go to sleep, a miracle happens. OK? And the problem that brings you in here is solved.

C: Hm hm.

SdeS: But you can't know it.

C: Mm hm.

SdeS: 'Cause it happens while you're sleeping.

C: OK.

SdeS: OK?

C: All right.

SdeS: So, when you wake up tomorrow morning, what will you notice, what will give you the clues that maybe a miracle has happened? (de Shazer, 1994:250)

When the client answers, it's important to listen for what the client will be doing differently, even though feelings are important. The goal, as Walter and Peller state it, is to elicit statements that "include some action, some behavior, some new framing, or something clients will say to themselves or others" (1992: 78).

At the training I led, I described these functions of the miracle question:

It helps to get a three or four dimensional picture of where the clients want to go;

It helps the clients believe that it is possible to have what they want; and

It helps the client begin to specify the behavioral cues from those around them which provide feedback that they are on the right path.

I demonstrated the miracle question with Debbie (not her name) who wanted help ending her smoking habit.

JP: Let's presume that when you go to sleep tonight, by some amazing miracle, or maybe not so amazing, when you wake up in the morning a miracle has occurred. And your life is transformed, you're breathing more freely, you have better health, you have all these things that are changed over. What would be the first thing that you would notice?

D: Probably that I didn't cough and that I could breathe.

JP: OK, so you'd notice your breathing. All right, and what else would you notice?

D: I hope that I, er, I would be able to run farther.

JP: Okay, so you would go for a jog and notice you'd be able to run? OK, what else?

D: Oh, I would have great breath.

JP: You'd have great breath?! (laughter) You mean it wouldn't smell? Sweet-smelling breath, oh, OK. OK, so you'd be able to breathe through the night, be able to run further and have sweet-smelling breath. Great, OK, what else? What would other people notice? You live with a group-live with a family?

D: Yeah.

JP: What would they notice?

D: I had more energy.


D: The house wasn't filled with smoke.

JP: OK, so the house would smell better.

D: Yeah,


D: I wouldn't be grumpy cause I quit smoking (laughs).

JP: You wouldn't be grumpy. So, if you're not grumpy, what are you? What would they see?

D: I would be fairly cheerful.

JP: Fairly cheerful, you wouldn't be just cheerful?

D: Well, not first thing in the morning (laughter).

JP: That's a different miracle.

D: Yeah...

I held parts of the solution while she explored her future and found more pieces of it. I also carefully restated some of them so that they represented the presence of something rather than an absence, as when I restated "The house wasn't filled with smoke" to "The house would smell better." Though trying to quit smoking can be a serious matter, Debbie humorously presented the balances and compromises she would face, as when she said she would be perceived as "fairly cheerful" in the morning.

At my presentation, I reminded the group that some clients might need help giving realistic answers to the miracle question. Usually, a casual acceptance of their first answer followed by a repetition of the hook of the miracle question serves to ground the client. For example, a client who answers that she won the lottery can be asked "And what would that allow you to do which would solve this problem?"

Five Useful Questions

de Jong and Miller (1995) have described five useful questions that support, reinforce, highlight, and buttress this pull of the future state on the client. The first one takes advantage of the spontaneous improvements clients sometimes make during the time between their making of their first appointment and the appointment, as in, "Anything better since you made the appointment?"

The second type of useful questions, exception finding questions, help clients to locate and appreciate moments in their past when the present problem got handled. For drinkers, this might be times when they went for a long period without drinking; for a couple that fights constantly this might be times when they got along well. The basic format is roughly, "Can you think of a time in the past [time period] that you did not have a problem with?" this can be followed up with questions like "What would have to happen for that to occur more often?" (Berg, 1992:no pagination)

Berg and Miller place the miracle question as the third type of useful questions. Their fourth type, scaling questions, ask clients to express their feelings about something on a scale ranging from 1 at the low end to 10 at the upper end. Berg calls it a "versatile, simple, and useful" tool which can be asked of anyone old enough to understand numbers. Scaling questions can help in many clinically difficult situations such as when a problem is vague or there are vast disagreements about issues, as might be the case in family situations (Berg, 1992:no pagination).

The fifth type of useful questions, coping questions, is useful when clients seem really discouraged and mired in their difficulties. It provides a way of "gently challenging the client's belief system and her feelings of hopelessness while, at the same time, orienting her toward a sense of a small measure of success" (Berg and Miller, 1992:89). The form of this question is simply to listen to the client's complaints and then ask, "With all of that going on, how do you manage to cope?"

In the ninety minute presentation I did, I had just a few minutes to briefly cover some of the five useful questions. We covered the miracle question, and discussed how to use exception-finding questions. Practitioners of SFT begin all their sessions after the first one with, "What's better?" (Berg, 1992). Formally, this parallels their first questions in the first session about any changes between the arranging of the appointment and the appointment itself.

SFT practitioners actively take control of the reins of their sessions. They are not holding a conversation, they are conducting therapy, and recognize the benefit to the client of the therapist keeping a firm hand on the reins of the process of what occurs within their shared therapeutic encounter. SFT recognizes that clients and therapists are engaged in a very real, but covert, struggle over who will control their encounter. This happens not only in therapy, but in other types of relationships as well.

When using the SFT approach the therapist takes charge right at the beginning. Getting that first question out fast, "What's changed since you called and made the appointment?" or "What's better?" begins to craft the session in a positive direction before the client realizes the struggle for the reins of the relationship has begun.

Powerful therapists like Milton Erickson have been accused of being overly concerned with interpersonal power and manipulation (Haley, 1994:56). But by assuming that power, and quickly establishing it as done in SFT, a therapist expedites the pace of therapy, that is, he or she keeps it brief.

Perhaps these five useful questions, in their simplicity, can be compared to a drum, which is a hide drawn and tied around a hollowed out piece of wood. But this simple instrument, in the hands of a master drummer, can really get people up and moving.

Reactions and Feedback

As a practice exercise, I asked participants to form groups of three and ask each other the miracle question. In the discussion that followed, I had to clarify that the miracle question was not asking people to believe in miracles, it just poses a question which allows the client to think of the ideal solution to their problem without all of the pesky details getting in the way. Haley (1994:56) states that the paradox of therapy is finding ways to induce change without activating the forces that prevent change. The miracle question addresses this paradox of thinking of change without triggering thoughts that nullify images of change.

Other responses from participants included:

It was very powerful. Just to see in people's gestures, how they relaxed or smiledIt was almost like taking them out of this setting and putting them into a different setting.

It provided hope and dreaming in that they could get excited about the future and the steps they were going to take.

I thought the question about the senses was very effective as another way to reach the person. We were helping our "client" do more exercise and asked her would it affect her sense of smell. Her response was, "Oh, well, yes! I could breathe better, I could smell better." I was the observer and had not thought about smell, and neither had she until it was asked.

"Client": It stopped me right in my tracks. It took me to another plane, where all of a sudden things were just coming in and I could really smell differently.

I also had a short evaluation form I asked people to fill out. I got 28 returned out of an original group of about 32 people. On a scale of 5 with 1 being low a low rating and 5 being high, I received a mean score of 4.6 for the usefulness of presentation. One participant left a message a week later:

Hi John, you don't know me, but I was at the training that you did on the Solution Frame, the Miracle Question. It was a great training. I wanted to tell you that. I got a lot out of it. I also wanted to tell you that as a result of our group asking each other the miracle question, I was so inspired that I went out and enrolled in three classes. Things that I always wanted to do, but am not doing. So now, I am doing a couple of art classes, an art history class, and probably a dance movement class. I just wanted to let you know that you are a good trainer and I got a lot out of your training.

The last feedback that I received about this training was in January 1996, sixteen months after my presentation. At a community meeting sat Debbie, the woman who volunteered for the demonstration of the miracle question. She told me how much she enjoyed that training, and that she had not had a cigarette since that day...Wow, a miracle!


Bateson, Gregory. (1979). Mind and Nature: A Necessary Unity. New York: Bantam.

Berg, Insoo Kim, and Miller, Scott D. (1992). Working with the Problem Drinker: A Solution-Focused Approach. New York : W.W. Norton.

Berg, Insoo Kim. (1992). Miracle Picture: A Vision of Solutions in Couple Therapy. In syllabus materials for the Special Problems in Marital and Couples Therapy: Dealing Effectively with Difficult Couples Conference. Portola Valley, CA: Institute for the Advancement of Human Behavior.

de Jong, Peter, and Miller, Scott D. (November 1995). How to Interview for Client's Strengths. Social Work, 40(6):729-736.

de Shazer, Steve. (1994). Words Were Originally Magic. New York: W.W. Norton.

Haley, Jay. (1994). Zen and the Art of Therapy. Family Therapy Networker, 1994, January/February: 55-60.

Haley, Jay. (1981). Development of a Theory: The History of a Research Project. In Reflections on Therapy and Other Essays. Washington, DC: The Family Therapy Institute, pp. 1-65.

Harper, Robert A. (1959). Psychoanalysis and Psychotherapy. Englewood Cliffs, NJ: Prentice-Hall.

Lipchik, Eve. (1994). The Rush to Be Brief. Family Therapy Networker, 1994, March/April:37-39.

Miller, Scott D., and Berg, Insoo Kim. (1995). The Miracle Method: A Radically New Approach to Problem Drinking. New York: Norton.

Walter, John L., and Peller, Jane E. (1992). Becoming Solution-Focused in Brief Therapy. New York: Brunner/Mazel.

Wegner, Daniel M. (1989). White Bears and Other Unwanted Thoughts: Suppression, Obsession, and the Psychology of Mental Control. New York: Penguin Books.