I just read the article and it seems that the authors are not singling out the therapeutic relationship itself but, rather, talking about the "facilitative conditions" of therapy which certainly include the therapeutic alliance, but also extend to the therapist's empathy, attunement to the client, and the congruence or fit between therapist and client. Even in therapies like CBT and DBT which do not focus on the therapeutic relationship, it's important for the T to be in tune with the client in order to understand what the client's issues are, where the client is at, what self-care strategies the client does and does not possess, when to push the client and when to give the client more assistance and so forth. In those instances, the client may have no need to discuss "the relationship," but the client will certainly make more progress if the T is attuned to the client, the T is addressing the correct issues with helpful strategies, and the T and the client are a good fit.

That seems different to me than therapy necessarily focusing on "the relationship." I'm certainly someone who needs that relationship in my own therapy. Like the OP, I'm in a type of therapy that focuses on that relationship. In fact, I'm in therapy because I grew up with neglect/abuse and never had a secure attachment. In order to heal from that, I need to go through the process of having a secure attachment with my T. I need that instead of things that others might need, like behavioral modification, mindfulness, self-coping skills, etc. I do fine with those things, but I need to learn how to heal, trust, and develop secure attachments in my relationships. On the other hand, I have friends who are in therapy and don't need the kind of T relationship that I do. They need an attuned and empathetic T, but what they struggle with are self-coping skills, destructive behaviors, bipolar episodes, etc. It's unnecessary for them to put time into contemplating or discussing the therapeutic relationship. The "fit" between client and T matters (they need a T with the right skill set), but "the relationship" is not a topic that needs to be analyzed. I tend to agree with Chris when she says that everyone who is in therapy needs something a little different, and it's great that there are so many different styles of therapy to accommodate those different needs. Sometimes, it also takes clients a few tries before they figure out which type of therapy works for them.

I also do not think that the percentage thing can be applied beyond the very specific examples and studies conducted in the article. It would be silly to say that, for any client in therapy, X percent of the outcome can be attributed to this factor, and X percent of the outcome to that factor. That's way too simplistic. In actual practice, those percentages would be different for each client. For me, the therapeutic relationship probably accounts for something like 80% of my progress (if I had to do something as arbitrary as attach a number to it). For my friend who also sees my T, the relationship probably accounts for somewhere around 10% of her progress. It's just not that important to her, or her mental health concerns. She's bipolar and in a volatile relationship, and she needs a T to help her learn how to calm herself down and not act-out or become destructive when she gets upset. Her therapy is focused on that. She does not feel any particular attachment to T; T is simply a tool to learn how to regulate her emotions and behaviors.