The question is a little false from the start. Somatic therapy isn’t an alternative to talking — you’ll talk for most of the hour. What changes is what the therapist does with what they notice, and what counts as evidence that something happened.
What a talk therapist tracks
Content, mostly, and its shape. What happened, what you made of it, what you keep doing, where the story contradicts itself. A good psychodynamic or CBT therapist is enormously attentive — to the thing you skipped, the word you picked, the belief sitting under the complaint. The bet is that if the material gets clear enough, and you look at it often enough, something moves.
What a somatic therapist adds
All of that, plus a running read on your physiology. They’re watching for the moment your breathing changes, and they treat that moment as at least as useful as the sentence you were in the middle of. Often they’ll stop the sentence to stay with it.
The underlying bet is different too. Some states have to finish in the body rather than be understood, and talking about an activated state while staying carefully outside it can be a very sophisticated way of avoiding it.
Where the difference actually matters
It matters most when insight has run out. If you can narrate your problem fluently and it hasn’t moved, more narration probably isn’t the missing piece. It matters for trauma that predates language, and for the kind that shows up as a physical reaction rather than a memory — the startle, the shutdown, the nausea with no thought attached.
It matters less than people think for ordinary trouble. Grief, a bad job, a decision you can’t make, a relationship going wrong. Those respond well to a thoughtful person paying attention, and the method matters far less than whether you trust them.
The honest comparison
Talk therapy has far more trials behind it, especially the structured forms. That matters, and it matters less than the phrase “evidence-based” implies — those trials mostly screened out complex patients, lost a lot of people to dropout, and often beat comparison treatments that were built to lose. Anyone telling you the research settles this, in either direction, is overselling it.
What the research doesn’t capture well is fit, and fit predicts outcome better than method does. A somatic therapist you find slightly embarrassing will help you less than a CBT therapist you trust. That’s not a reason to pick at random. It’s a reason not to agonise.
You don’t have to choose forever
Plenty of people do a stretch of one and a stretch of the other. Plenty of therapists work both ways depending on what the hour needs. Start where your instinct points, give it six to eight sessions, and notice whether anything is different — not whether you feel better every week, but whether anything outside the room has moved.