Most people choose a therapist the way they choose a dentist: a referral, an open slot, an insurance match. Then they spend months quietly wondering whether the problem is the therapy, or them.
Here is the uncomfortable part. Of all the things researchers have measured about what makes therapy work, one of the most consistent predictors of how you'll do is not the brand of treatment. It is the quality of the working relationship between you and the person across the room. And most of us never evaluate it, because evaluating it feels rude, ungrateful, or like proof that we're "difficult."
You are allowed to ask whether this is working. In fact, the research suggests it is one of the most useful questions you can ask.
What researchers mean by "the alliance"
The term is therapeutic alliance, and the most widely used definition comes from psychologist Edward Bordin. He described it as three parts working together:
- Goals: you and your therapist agree on what you're trying to change.
- Tasks: you both believe the activities in session (talking through memories, practicing skills, examining patterns) are a sensible way to get there.
- Bond: there is enough trust and warmth that you can say true things.
Notice what this is not. It isn't "do you like them?" or "do they make you feel good?" Plenty of people feel wonderful in a session that goes nowhere, and plenty do their hardest work with a therapist who isn't especially charming. The alliance is about whether you're a team pointed at the same target.
How strong is the evidence?
It's one of the most studied questions in psychotherapy. A large meta-analysis by Flückiger and colleagues, published in 2018 and covering roughly three hundred studies, found a moderate, reliable link between the quality of the alliance and treatment outcomes. The link held across different kinds of therapy.
Two cautions keep this honest. First, the association is moderate, not magical. A good alliance doesn't guarantee success, and a rocky start doesn't doom you. Second, correlation is tangled: people who are improving may rate the relationship higher, and a good relationship may help people improve. Researchers think both are happening. Still, the finding has been replicated enough that major professional reviews, including those from Norcross and Lambert's work on evidence-based relationships, treat the alliance as a core element of effective therapy rather than a pleasant extra.
The practical translation: fit isn't a luxury. It is part of the treatment.
The trap of judging fit by how a session feels
Here is where most people go wrong. They use the emotional temperature of one session as the verdict.
A hard session can mean your therapist is doing exactly the right thing. A warm, easy session can mean you're both avoiding the thing that matters. And on a bad day, your mind will happily recolor the whole relationship; people in low moods tend to recall more negative material, which can make a decent therapist look distant and a slow stretch look like failure.
So instead of asking "how did I feel today?", ask Bordin's three questions over a longer window.
Goals. Could you say, in a sentence, what you're working on? Could your therapist? If you've been going for two months and neither of you can name it, that's worth a conversation, not a verdict.
Tasks. Does the way you spend the hour make sense to you? You don't need to understand every technique. But you should be able to get a reasonable answer when you ask "why are we doing this?" A good therapist welcomes that question.
Bond. Can you say the embarrassing thing, or at least say that you're holding it back? Safety isn't the absence of discomfort. It's the sense that your discomfort will be met rather than managed away or judged.
Early signals are worth taking seriously, but not as a sentence
Research on early alliance suggests that how the relationship feels in the first handful of sessions carries information about later outcomes. That's useful, but it isn't a trapdoor. Alliances also repair. Psychologists Jeremy Safran and Christopher Muran have studied "alliance ruptures," the moments when you feel misunderstood, dismissed, or pushed too hard, and found that when a therapist notices and works through a rupture with the client, the relationship can end up stronger than before.
This means a bad moment is not a bad fit. What matters is what happens next. Does your therapist get curious when you say "that didn't land for me"? Or do they get defensive, or move on as if you hadn't spoken?
How they respond to criticism tells you more than how they respond to praise.
Signs it may be a genuine mismatch
Some patterns are worth taking seriously, especially if they persist across several weeks:
- You consistently leave feeling judged or smaller, not just challenged.
- Your concerns about the approach are brushed aside or interpreted away every time you raise them.
- You find yourself performing a more together version of yourself for the entire hour.
- You don't know what the plan is, and asking doesn't produce one.
- Something about their identity, background, or worldview makes you feel you'd have to explain basics before being understood, and you aren't willing to do that work.
None of these automatically means the therapist is bad. A mismatch is not a moral failing on either side. It simply means this pairing isn't giving the work what it needs.
Measuring it instead of guessing
One of the most interesting findings in this area comes from Michael Lambert's research on routine outcome monitoring. In studies where clients regularly filled out brief questionnaires about how they were doing and therapists saw the results, clients who were not improving as expected tended to do better than similar clients whose therapists received no feedback. The benefit was concentrated in the clients most at risk of getting worse. Therapists, it turns out, are not reliably good at noticing on their own when someone is stalling.
You can borrow the spirit of this without a clinic's software. Give your own experience a place to live outside the heat of any single session, and look at it over time.
Your next moves
- Write one sentence for each of Bordin's three parts after your next session. What are we working on (goal)? What did we do about it (task)? How safe did I feel being honest (bond), scored 1 to 10? Repeat for four sessions and look at the trend, not the single data point.
- Ask one direct question next session: "Can you tell me how you're thinking about what we're working on, and what you'd expect to see change?" Their answer is information about the goals and tasks.
- Try a small, honest criticism. Pick something minor, such as "I wish we'd spend less time on my week and more on my family," and watch how it is received. This is a low-stakes rupture test.
- Name what you're withholding. If there's something you haven't said because you doubt it will be received well, say that sentence out loud: "I've been holding something back because I'm not sure how you'll react." The reaction is the data.
- Set a review point. Tell yourself, and ideally your therapist, "Let's check in at session eight about whether this is working." A planned review turns the question from an accusation into routine maintenance.
- If it's still off after a repair attempt, try someone else. Consider a consultation call with one or two other therapists. Switching is not betrayal; it's the same thing the research says matters, applied to you.
Why it helps to keep your own record
The reason this question is so hard to answer is that therapy happens in 50-minute bursts separated by a week of life. By the time you're wondering whether it's working, the details are blurred and your mood is doing the narrating.
That's the gap Sesh was built for: a private place to put what came up, how you felt about the session itself, and what you meant to say but didn't, so that when you ask "is this the right fit?" you're answering from a record rather than a mood. If that sounds useful, you can take a look at https://sesh.lumenlabs.works. And if not, the one-sentence check after each session works fine on the back of an envelope.