Bond, goal and task agreement; how ruptures show up in session, how repair works, why the alliance is measured, and the cultural side of practising in Turkey.
The alliance is one of the most used and least defined concepts in clinical conversation. When someone says “the alliance is good” in supervision, what they usually mean is that the client attends regularly and looks comfortable talking. Both observations may be accurate and still say almost nothing about the alliance.
The clinical value of the concept appears when you turn it from an atmosphere into a structure. The alliance is not a feeling but a working relationship with three separate components, each of which can break down on its own. One component can be sound while another has collapsed — and more often than not, that is exactly the case.
Bordin’s now-classic framework splits the alliance into three parts: the bond, agreement on goals, and agreement on tasks. The practical use of this division is that it turns a vague observation like “the relationship isn’t going well” into three separate, answerable questions.
The emotional bond: the dimension of trust, acceptance and warmth. Whether the client feels unjudged, and whether they feel safe enough to say something shameful.
The bond is the most talked-about component of the alliance and the one that causes the least trouble. In a great many cases it forms at a reasonable level and stays there. Most clinical difficulty happens in the other two components — but because the bond looks sound, the alliance is assumed to be fine.
The typical sign of a weak bond is staying abstract in session: the client talks about events but not about themselves, describes third parties, changes the subject when feeling comes up, and keeps looking for your approval.
A shared answer to the question of what this work is trying to change. It sounds like the simplest component and in practice it is the one most often missed.
A common example in practice in Turkey: a client comes saying “I can’t sleep”; within a few sessions the therapist sees an ongoing relational impasse underneath the sleep problem and moves the focus there. Clinically the shift may well be right. But if it has not been discussed openly, at the eighth session the client still thinks they came about their sleep, and the question “what am I getting out of these conversations” is quietly growing in their mind.
Signs that goal agreement has broken down: the client asking “what are we going to talk about today” at the start of the session, never returning to last week’s topic, asking more and more often when the work will finish.
Finding the things done to get to the goal meaningful. Homework, an exposure hierarchy, keeping a diary, the empty chair, breathing work, free association — whatever the method, the client has to understand why it is being done and find it reasonable.
Task agreement is the component that most often breaks down quietly. The client does not do the homework but says “I couldn’t, I was busy”. That sentence is usually not a time-management problem but the polite form of “I didn’t understand what this was going to do for me”.
A habit worth keeping: after setting a task, ask the client to give the rationale for it in their own words. Your having explained it does not mean it was understood.
The idea that the alliance is built in the first sessions and then stays fixed is widespread and wrong. The alliance fluctuates throughout, and the fluctuation is itself clinical information.
A few patterns to expect:
Mistaking these fluctuations for breakdown makes the therapist back off unnecessarily. Never noticing them lets a rupture deepen quietly.
The clinical literature describes ruptures in two basic forms, and the distinction genuinely helps in session.
Withdrawal. The client moves away from the relationship without letting it show. Answers get shorter, “I don’t know” comes up more often, the conversation gets shallower, an over-compliant manner appears. A client who agrees with everything you say is usually in a harder position than one who agrees with nothing. Arriving late, cancelling sessions, and the important sentence said at the door on the way out all belong in this group.
Confrontation. The more visible, more uncomfortable and — paradoxically — easier form to work with. The client objects directly: “what is the point of that question”, “what you said last week didn’t fit me”, “we talk this much and nothing changes”. Tension arising over fees, appointment times or the cancellation policy usually belongs in this category too.
Withdrawal ruptures are more frequent and more dangerous, because they go unnoticed. In practice most therapists recognise confrontation as a rupture but wave withdrawal away as “the client is tired today”.
There are signals from your own side too, and they are no less informative than the client’s: a slight reluctance before the session, doing too much of the talking during it, explaining yourself constantly, checking the clock more often, putting off writing the note afterwards. None of these says anything on its own; when they appear together, there is something to look at.
A rupture is not in itself a bad thing. What is clinically decisive is whether it gets repaired — and a repaired rupture often leaves a more valuable experience behind than a course of therapy with no rupture in it at all. For many people, the consulting room is the first place they experience something going wrong in a relationship, being talked about, and being put right.
Repair has an order that works.
Notice it and stop. Passing over the rupture and carrying on with the agenda is the easiest option and the most expensive one. The moment you notice that something has changed, whatever is being discussed, you have to stop.
Name it, do not interpret it. There is a large difference between “I noticed the tone of the conversation changed in the last few minutes” and “I think you’re angry with me”. The first offers an observation and leaves the client room; the second imposes an interpretation and is usually met with denial.
Do not get defensive. This is the hardest part of repair. When a client says “you didn’t understand me”, the reflex to explain is strong — and explaining almost always deepens the rupture. Understanding comes before being right.
Own your share. Where the therapist really does have a share, saying so does not weaken the alliance; it does the opposite. “I made that interpretation too early last week” is not a loss of authority but evidence that the relationship is real.
Go back to the goal or the task if you need to. A good share of ruptures are structural rather than emotional: the goal has drifted, or the task was not understood. In that case repair is not an emotional conversation but an explicit renegotiation. “Let’s get clear together again about where we’re going” is, in most cases, more effective than a transference interpretation.
Talk about the repair itself. Once a rupture has been repaired, naming together what happened consolidates the work: “Something went wrong last week, we talked about it, and it got better. How does that go in your relationships outside this room?” That question is the bridge that carries the experience in the room out of it.
The mismatch between the therapist’s rating of the alliance and the client’s is one of the field’s most consistent observations. Therapists struggle to predict the client’s experience from their own impressions, and the direction of that struggle is generally towards optimism.
The reason is simple: given the balance of power in the relationship, it is not easy for a client to tell their therapist directly that they are dissatisfied. In a context where deference to the expert is culturally strong, as it is in Turkey, saying “this isn’t doing me any good” is a serious burden for a client to carry. One way of making that sentence easier to say is to take it out of the conversation and put it on a form.
Two approaches are used in practice:
Detailed alliance measures. The Working Alliance Inventory and its derivatives assess the bond, goal and task dimensions separately. Because they show which of the three components is weak, they feed directly into the formulation. They are not short enough to administer every session; using them a few times across the work — at the fourth and twelfth sessions, say — is an arrangement that works.
Very short end-of-session measures. Instruments of a few items, such as the Session Rating Scale, are completed in a minute at the end of each session. They give no detail, but their timing advantage is considerable: you can take a dip up at the start of the next session.
The real benefit of measuring is not the number but the conversation the number opens. When you see a low mark on a form completed at the end of a session, the following week you can say: “it looks as though something was missing here last week — shall we talk about what it was?” That sentence is often more repairing than everything the measure was measuring.
One warning: turning alliance measurement into a tool for tracking therapist performance renders the tool useless in short order. The client softens the form to protect their therapist, and the therapist starts overlooking the low-scoring sessions. Measurement exists so that the therapist can see their own work.
That an alliance can be built in remote sessions is by now largely settled in the field. But the way it is built is not the same, and knowing the differences makes online work easier.
Bodily information narrows. On screen you see from the chest up. A jiggling foot, clenched hands, leaning back in the chair — some of the signals that carry the most information in a session never enter the frame. The way to compensate is to ask: “what is happening in your body right now?” is needed more often online than face to face.
The illusion of eye contact. Look at the camera and you are looking at the client but not seeing them; look at them and they see you looking somewhere else. In the first sessions this can produce a mutual sense of not being listened to. Saying so once, openly, is enough to settle it.
The fragility of the frame. A room supplies a frame of its own; on screen the two of you have to build that frame together. A closed door, headphones, notifications off, staying in the same place for the whole session, not joining from the car or while walking. These are not technical details but the material ground of the alliance.
What interruptions mean. If what to do when the connection drops is not agreed at the outset, each interruption accumulates as a small experience of being abandoned. A simple agreement is enough: if we lose the connection, we try to reconnect within two minutes, and if that fails I will call you.
Privacy. For many clients in Turkey the biggest obstacle is not bandwidth but finding a room at home where they can be alone. Sessions done with headphones, by typing, sometimes from a car, usually have this behind them. Knowing where the client is and who else is at home is a precondition of understanding how far they can open up in session.
Crisis situations. Working remotely, you need to record from the outset where the client physically is and who can be reached in an emergency. This is not a formality that damages the alliance; discussed calmly in the first session, it is a frame that shows the client they are being taken seriously.
The concept of the alliance was largely developed in the Western clinical literature, and some of the assumptions it carries — individual autonomy, direct communication, the expectation of an egalitarian relationship — do not hold for every client in Turkey. That is not a deficiency but a difference to attend to.
The expectation of an expert. “Hocam, you tell me — what should I do?” is something most therapists in Turkey hear several times a week. (Hocam, literally “my teacher”, is the ordinary respectful way of addressing a professional.) It is easy to read this as resistance or a lack of autonomy, and that reading is usually wrong. The cultural template for a relationship with a health professional is one in which the expert supplies the knowledge. A client is expected to arrive carrying that template.
The approach that works is not to refuse the expectation but to name it and shift it gradually. Being more directive in the early sessions and handing the space back to the client as the work goes on does not damage the alliance; it builds it. A course of therapy that opens with “I don’t tell people what to do” feels, to a great many clients, like being abandoned.
The presence of the family. Individual goals frequently collide with family expectations. What a client gains by learning to set a boundary can be experienced as a crisis within the family system. Not allowing for this when agreeing goals leaves the client alone with a cost they cannot carry. The goal is not “set a boundary with your mother” but “what change in your relationship with your mother would be bearable for you”.
Stigma. There are still settings where going to therapy is something people hide. An appointment arranged by slipping out of work, the invoice sent to an email address a spouse cannot see, a request that the session reminder arrive without a name on it — all of these are part of the alliance. A client’s need for secrecy has to be treated as a condition of the work, not as a logistical detail.
Forms of address and distance. Turkish separates the formal siz from the informal sen; choosing between them, deciding whether to accept being called hocam, deciding whether to address the client by first name or by surname — these look like small decisions, and they set the warmth and the hierarchy of the relationship directly. There is no single right answer; what matters is that the choice is made deliberately and held consistently throughout.
Contact outside sessions. Contact through messaging apps is the area that produces the most boundary trouble in practice in Turkey. Replying to a long message that arrives at midnight looks in the short term as though it strengthens the alliance; in the medium term it both exhausts the therapist and reduces the importance of the session. The rule has to be set at the outset, explicitly and with its reasoning: which channel, which hours, what kind of content, and how quickly a reply comes.
Gifts, hospitality, being a guest. Arriving at a first session with a box of chocolates, sending greetings at a religious holiday, insisting on tea at the end of a session — for many clients in Turkey these are part of the code of politeness, not an attempt to breach a boundary. A flat refusal can cool the relationship. The clinical judgement belongs in the value of the gift and in how often it repeats, not in the fact of it.
A religious frame. Clients often describe their distress in religious language: sabır (patient endurance), imtihan (a trial one is being put through), kader (fate). Trying to correct that language damages the alliance fast. Working from inside the same language is usually possible and more effective.
Multiple relationships in small towns. Outside the big cities it is ordinary for a client to be your neighbour, your child’s teacher or your spouse’s colleague. You have to find a path between the fact that this cannot always be avoided and the need to discuss at the outset how boundaries will be held in those conditions. Supervision has a particular place here.
Thinking of the alliance as separate from technique — as a “preparing the ground” phase that comes before it — is a common mistake. In practice the alliance is built, or broken, precisely while technique is being applied.
A well-timed interpretation deepens the bond. The same interpretation made too early produces a rupture. An exposure exercise whose rationale the client understands strengthens task agreement; the same exercise without that understanding starts a silent withdrawal. Sharing the case formulation with the client is the most direct way of establishing goal agreement; keeping it to yourself is the most common cause of goal drift.
So it is more useful to think of the alliance not as a precondition but as a standing checkpoint. Three questions at the end of each session are enough:
Writing down the answers to those three questions for weeks on end is the cheapest way of noticing where the alliance is weakening. Taken into supervision, the answers make the discussion run on a specific component rather than on impressions — and that gets you a great deal further than “the alliance looks good”.
The alliance is not something that runs by itself in the background of the work; it is a working relationship rebuilt in every session, able to break and able to be repaired. Tracking it through its three components makes it possible both to notice a rupture early and to aim the repair.
The client being able to complete a short rating form at the end of a session is the lowest-cost form of that tracking — and in practice it is the only outside perspective the therapist can add to their own impressions.
Let's talk through how this would work in your own practice.Book an intro call
In about fifteen minutes we'll walk you through the product on a demo built around your practice, and answer whatever is on your mind.
30-day money-back guarantee.