The Centre for Solution Focused Practice

Endings: a personal journey


The Solution Focused Listserv set up by Harry Korman has done so much for the development of the model, supporting the teasing out of difference as well as providing an unfailingly reliable source for references, referrals and clarifications within the field. Just recently the listserv has been focusing on session endings and in particular the role of tasks and homework. This has prompted me to thinkabout my personal development, the shifts and changes in my practice as regards endings over the past 35 years. I would wish to add that this is my own, Evan’s, personal journey, since I think that Chris Harvey and I, the members of the BRIEF team, have ended up in different places regarding endings and I doubt that I can do justice to Chris and Harvey’s positions.

In 1987 when we started exploring Solution Focused Brief Therapy we started off attempting to do what Steve de Shazer and the Milwaukee team seemed to us to be doing, as far as we were able and as far as we understood what they were doing, and in those early days session endings seemed to be central to the model. The questions we were asking seemed designed, substantially, to help us to establish what task to give at the end of the session and so during the session we were focused on findings things out. Were there exceptions? Were the exceptions deliberate or spontaneous? What alternative possibilities for action arose in the client’s response to the miracle question? And crucially what was the nature of the client-worker relationship? Was it a visitor or a complainant or a customer type relationship? Finding answers to these questions shaped the sort of tasks that we gave clients which were heavily influenced by de Shazer’s thinking about ‘skeleton tasks’ in Keys to Solution in Brief Therapy (1985). Within this ‘model’ change occurred as the result of the client behaving differently out of the session and the key to this different behaviour was the task which provided a gateway to the client either doing something different, or perceiving things differently or shifting their expectation of change occurring all or any of which might be useful. Thus the task, or ‘experiment’ as de Shazer often described tasks as, was intended to promote difference and, as he writes ‘since solutions are not predictable in any detail and since there is more than one potential way of behaving in the future without the complaint, the new set of expectations can be constructed out of any satisfactory or beneficial changes. Any change stands a chance of starting a ripple effect, which will lead to a more satisfactory future. Therefore, the brief therapist reacts to any change as an indication that things are starting to go right for the clients. It does not seem to matter if a particular change is new or different behaviour, or if it is an exception to the rules of the complaint, or even if it seemingly has nothing to do with the complaint. Any change is a difference that could well prove to be different enough to be part of the solution. In any case, any change can become part of the construction of a new set of expectations that will be part of creating the solution.’ (p 77) The striking repetition of the word ‘any’ on this page emphasises that de Shazer viewed tasks (or experiments) as vehicles for the emergence of difference and that it was difference rather than any specific difference that was important in his thinking.

Given that tasks were so central at this stage of SFBT’s development it was also understandable that the Milwaukee team were interested in maximising the receptibility of the task and this was achieved through the giving of ‘compliments’. Returning to Keys (1985) de Shazer writes ‘to facilitate and promote the introduction of the therapeutic suggestion, the therapist begins the message with compliments or statements about what, in the therapist’s opinion, the client is doing that is good for him or her. These comments may or may not have anything at all to do with the complaint. The purpose of the compliments is to build a “yes set” (Erickson et al., 1976; Erikson and Rossi, 1979; de Shazer, 1982a) that helps to get the client into a frame of mind to accept something new - the therapeutic task or directive’ (p91). Compliments were conceived, as the text makes clear, within a hypnotherapeutic frame.

Following the relative clarity of this early position, it was not absolutely clear in my view and contained a number of diverse strands, our thinking began to change in a number of ways. We began to see the conversational process during the session as the key to change, asking questions which invited the client into, or co-constructed, an alternative story or narrative or ‘reality’. We abandoned the visitor, complainant, customer relationship types distinction choosing to assume that everyone who comes to see us wants something and the only question is to frame what they wanted in a way that fitted with the Solution Focused approach. We were thinking more and more about de Shazer’s idea of ‘co-operating’ and where that might take us and began to be unhappy with thinking derived from hypnotherapy and we were wanting more and more to centre the client’s voice rather than our own. (I remember someone saying to me, having observed a session of mine, ‘during the course of your whole session it is the client’s voice that is central and then in the ending sequence you take centre stage’. I remember both agreeing and feeling uncomfortable with this.) So the first step was changing our language in a way that better supported our intentions. I remember deciding not to use the work ‘compliment’ and not to thinking about ‘tasks or homework’ and so I switched to ‘Solution Focused Summary’ and ‘Suggestions’. The Solution Focused Summary replaced the compliments and was a precis of the things that the client had said, using their own words as far as possible, that fitted with the idea ‘I can change’. In this way the client’s own voice was centred and the language would be derived directly from the client. And instead of ‘task or homework’ both of which seemed to me to have hierarchical implications – after all just think about who gives homework to whom and what the consequences tend to be if homework is not completed! – we had the idea of ‘offering a suggestion’. This shift in description helped to support our chosen assumption that every client arrives motivated for something, because if the client did not do anything (obvious) with the suggestion that we offered, then we were more likely to turn our gaze onto the suggestion and to question ourselves in relation to why it did not appear to have been useful rather than questioning or doubting the client. As it happened these shifts more or less coincided with our choice to abandon most of the skeleton task formulations that de Shazer offered us in Keys (1985) and pretty well always to offer the suggestion that the client might like to watch out for change, or watch out for the things that the client was pleased to notice or those that were taking the client in their preferred direction. This shift to always offering the same suggestion, fundamentally, meant that we no long wasted time, in our view of course, trying to work out what task to give at the end of the session, thereby taking time away from developing the conversation which by then we believed was more significant. These changes coincided with an overall reduction in average number of sessions whilst outcomes seemed unchanged. The idea that I was ‘offering’ a suggestion was important to me; the offering was invariably tentative rather than in any sense persuasive so that the client would reach out and take the suggestion from me rather than being instructed. By now of course the break has disappeared altogether, since the purpose of the break was to figure out the compliments and the task and to create (in a hypnotic sort of way) what was called ‘heightened attention’.

My current practice is perhaps a development of these intermediate changes and so a typical ending sequence for me these days is made up of 6 elements.

1. ‘Has anything been missed out or is there anything that you had in mind to say that I have not given you the opportunity to say?’. Virtually everyone says no. I sometimes ask ‘have we been talking about the right things?’ and virtually everyone says yes.

2. ‘What have you heard yourself saying today that might be worth remembering and taking back into your life with you?’ I often add that ‘it is of course difficult to know for sure what will end up being useful’. People are typically unpredictable in relation to what they select.

3. ‘Would you mind me adding a couple of things that you said that stood out for me?’ People say that that would be fine and I feedback in their language a couple of things that fit with the idea ‘I can change’.

4 & 5. ‘Do you think that it would be useful to make another time to talk now or would you prefer to get in touch with me when another session would seem useful?’ If the client says ‘let’s fix a time’ then I ask them when would seem useful.

6. And if the client does say that they want another time they I offer the client a friendly heads-up or warning ‘When you come back the very first question that I am going to ask you is ‘what’s been better?’ so if you start watching out for that now it might make your next session easier for you but of course what you do is up to you and it is not for me to boss you about’.

If the client does not make an appointment then I might end with ’something that is useful for lots of people is to watch out carefully for evidence of (further) change – sometimes it can be hard to notice change if we are not looking out for it and noticing it seems to make a difference for lots of people. Good luck!’.

My current practice is different both from Harvey’s and Chris’s. Harvey is I think, currently closer to de Shazer in his thinking about session endings. Chris on the other hand would probably say that I am not sufficiently trusting of the process, of the power (he would probably not use that word) of description and that I could drop the entire ending sequence apart from making another appointment (or not). Chris’ view might well be that the ending is to do with me and that the client’s time is being wasted – a cardinal sin!!!! Of course my view is that clients like an ending!

For anyone who would like to sign up to the Solution focused Listserv then you can connect at

de Shazer, Steve (1985) Keys to Solution in Brief Therapy. New York: Norton

Evan George


08 January 2023


Featured Video

What is SF - a 2020 version of the approach


July 9, 2020