Does talking therapy work? What actually makes a difference

“Does talking therapy work?” is a question I get asked a lot, usually within the first five minutes of a consultation call, with a slight edge of scepticism. 

It's a fair question. 

If you strip counselling back to its most basic mechanics, it does look, from the outside, faintly absurd: two people, sitting in a room (or, in my case, more often than not, walking through a wood in a walk-and-talk session), one of them talking, the other listening. That's it. That's the whole apparatus. How is that supposed to do anything?

Here's my honest answer: it isn't. Not directly. The talking itself isn't what changes you.

What talking does do is build something. And it's that something – not the words themselves – where the actual work gets done.

What makes talking therapy work?

Think of it like an iceberg. What you can see above the waterline – the conversation, the working through of a difficult week or a decades-old wound – is real, and it matters. But it's a fraction of the mass. Beneath it is the much larger structure holding the whole thing up: trust, built session by session, the experience of being met without flinching, disagreed with without punishment, understood on a bad day as readily as a good one. That's the part doing the work. The talking is just how you get access to it.

What does the research say?

This isn't just something I've noticed anecdotally. It's one of the more consistently replicated findings in the entire field of psychotherapy research, going back nearly a century.

In the 1930s, a researcher named Saul Rosenzweig noticed something odd: different schools of therapy, built on wildly different theories about what was wrong with people and how to fix it, kept producing broadly similar outcomes.

He borrowed a line from Lewis Carroll to describe it (“everybody has won, and all must have prizes”), an idea that later became known by the slightly less poetic name, the ‘dodo bird verdict.’ 

So, if wildly different techniques produce broadly similar results, something other than the technique is doing a lot of the heavy lifting. 

Researchers like Bruce Wampold, who has spent much of his career studying exactly what that ‘something’ is, argue it's largely the relational context in which the technique is delivered: the quality of the bond, the sense of being on the same team, working toward something you both believe in.

Carl Rogers got there even earlier, in his own way. In 1957, he argued that qualities including a therapist being genuinely empathic, authentic and accepting of the client weren't just nice to have alongside the “real” clinical work, but were among the conditions needed for therapeutic change.

Decades of subsequent research into what's now called the therapeutic alliance – essentially, the strength of the working bond between client and therapist – has tended to back this up: alliance quality predicts outcome about as reliably as anything we've found, often more reliably than which particular model the therapist is trained in.

Why does the therapeutic relationship matter?

What we're still working out is why, at the level of what's actually happening inside us. There's an honest gap here: we don't yet have a clear, settled explanation for the exact mechanism. But the clues we do have point somewhere interesting.

Work on co-regulation suggests that our nervous systems are, to some extent, always reading the safety of the people around us, and settling or activating accordingly; being in the consistent presence of someone attuned to you appears to have a measurable calming effect on the body's threat responses, quite apart from anything either of you says. It's not a full explanation, but it is the beginning of one.

And really, when you step back, none of this should be surprising. We are, before we are anything else, relational beings. We don't develop a sense of who we are in isolation; we develop it in the mirror of other people, from infancy onward, and much of what goes wrong for us psychologically – shame, mistrust, the conviction that we're somehow too much or not enough – takes root in relationships that failed to hold us well.

It would be strange, when you think about it, if the repair for that didn't also happen relationally. We were wounded in relationship. It makes sense that we'd heal there too.

You don't need research to understand the practical upshots of this either. Think about the difference between working for someone who has never once made you feel seen or valued, and working for someone who has. You already know which one you'd walk further for, try harder for, take a risk you wouldn't otherwise have taken for.

Being genuinely seen, heard and valued by another person doesn't just feel nice – it seems to unlock something in us: a willingness to do the harder, riskier, more honest thing, because we're doing it inside a relationship we trust to hold us if it goes wrong.

That, as far as I can tell, is what's really happening in the room.

Not two people talking. Two people building, slowly and deliberately, a relationship sturdy enough to hold what you've been carrying alone – and it's that relationship, not the conversation itself, that does the moving. The talking is just how you get there.