What to Tell Your Therapist When They Ask How You've Been

The question is harder than it sounds

"So, how have things been?" is not small talk. It is the intake for the entire session. Whatever you say in the next two minutes sets the agenda: it decides what gets explored, what gets skipped, and what your therapist writes down as the current state of things.

And almost everyone answers it badly. Not because they are careless or unmotivated, but because the question asks you to do something human memory is genuinely bad at: summarise a long, uneven stretch of ordinary time, fairly, from cold recall, under mild social pressure, in a room where you want to seem like you are doing okay.

Why your memory gives you the wrong answer

Two well-documented effects work against you, and neither is a personal failing.

Recency bias.The last two or three days get to stand in for the whole period. If you slept badly last night, the fortnight was bad. If this morning went well, the fortnight was fine. The other eleven days quietly don't vote.

The peak-end rule.Daniel Kahneman's work showed that we judge past experiences overwhelmingly by their most intense moment and how they ended, not by their average or their duration. So a fortnight with one very bad Tuesday and thirteen unremarkable days is remembered as a bad fortnight. A fortnight with one lovely Sunday at the end is remembered as fine, even if the preceding twelve days were flat.

Between them, these two effects delete exactly the thing that is most clinically useful: the ordinary days. The texture. The slow drift. Anyone can remember the crisis. Almost nobody can remember the eleven days of mild, sub-threshold flatness that preceded it, and that is usually where the actual pattern lives.

What a useful answer contains

You are not trying to deliver a symptom inventory. You are trying to give your therapist the shape of the period so they can spend the hour on what matters instead of excavating it. A useful answer has four parts:

1. The rough proportion. "About four of the last fourteen days were genuinely hard." A number, even an approximate one, is far more informative than "not great."
2. Where they fell. Clustered or scattered? Three bad days in a row is a different phenomenon from three bad days spread across a fortnight, and it points somewhere different.
3. What was around them. What else was true on those days. Sleep, workload, contact with particular people, alcohol, cancelled plans, the week a deadline landed.
4. What changed since last time. Whether the thing you agreed to try actually happened, and what followed if it did.

That is a two-minute answer that buys back twenty minutes of session.

Three ways to get there

1. Write three lines the night before

The lowest-effort option, and much better than nothing. The night before your appointment, write down: the hardest moment since last session, the best one, and the question you most want to ask. It will still be recall-biased, but it forces at least one deliberate pass over the period rather than an improvised one in the chair.

2. Keep a short running note

A note on your phone that you add a line to whenever something notable happens. This solves the recall problem for events, though it tends to under-record ordinary days, because nothing feels notable enough to write on a flat Wednesday. Flat Wednesdays are often the data.

3. Track a few numbers daily

The most reliable option, and the one clinicians consistently prefer, because it captures the ordinary days that the other two methods drop. Rate a handful of things once a day, mood, energy, focus, sleep, stress, and after two weeks you have something no memory can produce: a curve, with the bad stretch visible in it and whatever was happening around it sitting alongside.

The critical constraint is that it must be short enough to survive a bad week. A daily journaling practice that takes fifteen minutes gets abandoned inside a month, and patchy data is genuinely harder to interpret than no data. A few seconds a day is what actually sustains, and consistency matters far more than depth here.

What this changes in the room

When you can say "here are the last four weeks, the dip starts around the 14th, and that is the week I stopped sleeping properly," the conversation starts at a different altitude. You are not establishing basic facts any more. You are interpreting them together, which is the part you are actually paying for.

It also protects you from your own good days. Plenty of people arrive at a session on a decent morning, report that things are fine, and leave without addressing the fortnight that was not fine. A record does not have good mornings.

And if you are working toward an assessment or a medication review, the difference is larger still. "I think I've been more tired lately" and "my energy averaged 4/10 this month against 6/10 last month, and it tracks my sleep almost exactly" are not the same statement, and they do not get the same response.

One thing worth saying plainly

None of this is about arriving with a diagnosis, or turning yourself into a spreadsheet. Tracking does not interpret anything, and it is not a substitute for the clinical judgment of the person across from you. It just removes the guesswork from the first two minutes, so the remaining fifty-eight are spent on the work.

Your therapist is trying to see a pattern through a keyhole, one hour at a time, using your memory as the lens. Anything that makes that lens less distorted makes their job, and your outcome, better.