Imagine sitting across from someone for months while you quietly get worse, and neither of you says so. You nod. You say "fine, I guess." They nod back, kind and attentive, sure the work is going well. The research on this is uncomfortable. Therapists are surprisingly poor at noticing when a client is deteriorating.

That isn't an accusation. It describes how human judgment works when it can't see what's happening on the other side of the room. Understanding it can change how you use your own therapy.

The blind spot nobody advertises

Michael Lambert and his colleagues at Brigham Young University spent decades studying treatment outcomes. One finding recurs across their work: a meaningful minority of clients get worse during therapy, and clinicians tend not to predict which ones. In studies where therapists were asked to flag clients likely to deteriorate, they caught very few of the clients who actually did, even when they knew to look for it.

A separate line of survey research found that clinicians tend to rate their own effectiveness generously. Most placed themselves above average, which can't be true of most people. Optimism is probably useful for a therapist, because it sustains the work. It also makes them a poor judge of their own results.

The cause isn't carelessness. Therapy happens in a closed room, and a lot of the evidence about how it's going never enters it. You smile because you don't want to disappoint them. You skip the session where you felt awful. You tell them the week was "okay" because the real answer would take too long. Your therapist is working from a heavily edited version of you.

What happens when feedback gets added

Lambert's group tested a simple fix. Clients filled out a short questionnaire before each session (the best known is the Outcome Questionnaire, or OQ-45). If a client's scores showed they were off track, the therapist got an alert.

In these studies, clients flagged as at risk of failing did better when their therapists received that feedback than when they didn't. Rates of deterioration fell, and more of those clients reached meaningful improvement. Later meta-analyses of routine outcome monitoring have found a small to moderate benefit overall. The effect is concentrated in the clients who were struggling, which is where you'd want it.

Scott Miller and colleagues developed a lighter version used in what's called feedback-informed treatment. It has two very short forms:

  • The Outcome Rating Scale asks how you've been doing in four areas: personal well-being, close relationships, work or school, and life overall.
  • The Session Rating Scale asks how the session itself went: whether you felt heard, whether you worked on what you wanted, whether the approach fit, and whether the session felt right overall.

The whole thing takes about two minutes. The Session Rating Scale is built on the therapeutic alliance, the bond and agreement on goals between you and your therapist. That alliance is one of the more reliable predictors of how therapy turns out.

Why a number can say what you can't

There's a human reason this works. Saying "I don't think this is helping" to someone's face is hard. Circling 4 out of 10 on a line is easier. A rating gives you a way to be honest that doesn't feel like a confrontation, and it hands your therapist something specific to respond to.

It also fights a problem other writers on this blog have covered: your mood on the day colors how you remember everything. A rating made right after the session, or at the end of the week, is a record you can't revise later. When you look back over a month of scores, you see the trend instead of whatever today's mood says about it.

There's a caution here. Numbers are blunt. A low score doesn't mean therapy has failed. Hard sessions can score badly and still do real work, and a pleasant session can accomplish nothing. A score is a conversation starter. It isn't a verdict.

What to do if your therapist doesn't use measures

Many therapists don't. You can still run a version of this yourself, and nothing requires their permission. The part that matters is bringing it into the room.

A rating that stays in your phone does little. One you say out loud changes what happens next. "I rated this week a 3 and I think it's because I felt rushed at the end" gives your therapist something they could not have seen from where they sit.

If you say something like this and your therapist responds with defensiveness, that tells you something too. Most therapists welcome it. Research on alliance ruptures suggests that when a client raises a problem and the therapist handles it well, the relationship often gets stronger afterward.

Your next moves

  • Pick four life areas and rate them today. Use the Outcome Rating Scale's categories: you personally, your close relationships, work or school, and life overall. Give each a number from 0 to 10. Write the date next to them.
  • Rate every session within ten minutes of leaving. Ask four things: Did I feel heard? Did we work on what I came for? Did the approach make sense to me? Was anything missing? A 1-to-10 score for each is enough.
  • Look at the trend every four weeks. Check whether your life-area scores moved, not just whether sessions felt good. If they've sat flat or slipped for a month, write that down to bring up.
  • Name one low score out loud at your next session. Choose the lowest number from your ratings and say it plainly: "I gave last week's session a 4. Here's why." Then ask what could change.
  • Ask your therapist if they'd like to review your ratings together now and then. Some will say yes with enthusiasm. If they decline, that's information about whether this is the right fit.

Where this leaves you

It's tempting to treat therapy as something done to you by a professional who knows how it's going. The evidence suggests the opposite. The clearest picture of whether it's working is in your hands, and your therapist often can't see it unless you show them.

Sesh was built around that idea: that what happens in therapy shouldn't stay in therapy. It's a private place to keep your own record between sessions, including how each one felt and what shifted afterward, so you can walk back in with specifics instead of a vague "fine." If that sounds useful, you can look at it at sesh.lumenlabs.works. The two-minute rating works just as well on paper.