You can tell before they sit down. Something in the way they take off their coat a half-second slower, the apology already forming. "So, I didn't really get to the thing." You smile and say it's fine, because it is fine, and then you spend the first ten minutes of a session you both paid for, in different currencies, quietly re-teaching last week.
Here is the part nobody says out loud: it doesn't feel like their failure. It feels like yours. If you were better at this, they'd have done it. If you'd explained it more clearly, or been more inspiring, or chosen a client who was more serious. Working alone, there's no colleague to shrug at across the hallway and say "clients, right?" There's just you, the gap between what they promised and what they did, and the uncomfortable sense that the gap is a verdict on you.
It isn't. It's one of the most reliably documented phenomena in behavioral science, and once you see it clearly, you'll stop taking it personally and start closing it.
The gap has a name
Psychologists call it the intention-behavior gap, and it is enormous. In study after study, people who sincerely intend to do something, who are motivated, informed, and have told someone else their plan, routinely fail to do it. Reviews of the research led by the psychologist Paschal Sheeran found that intentions explain only a modest share of what people actually do, and that even when an intervention successfully strengthens someone's intention, the change in behavior that follows is noticeably smaller.
Read that again from the provider's chair. Your client's enthusiasm at the end of the session is real. Their "yes, absolutely, three times this week" is sincere. And it is still, statistically, a weak predictor of what happens on Thursday night when they're tired and the couch is right there.
This matters because most providers respond to the gap by working on the intention. We explain the benefits again. We get more animated. We ask them to commit more firmly. But intention was never the bottleneck. They already wanted it. What they lacked was something much smaller and much more mechanical.
Why the room lies to you
Part of the problem is that a session is a particular kind of place. The client is focused, supported, and sitting across from someone they want to impress. George Loewenstein, a behavioral economist at Carnegie Mellon, described what he called the hot-cold empathy gap: when we're in one emotional state, we are startlingly bad at predicting how we'll behave in another. In the "hot" state of the session, the client genuinely cannot imagine skipping the stretches. In the "cold" state of Thursday night, they genuinely cannot remember why they cared.
Neither version is lying. They're just two different people, and the one who made the promise isn't the one who has to keep it.
There's a second distortion, and this one is yours. When someone doesn't follow through, we tend to explain it with their character: not committed, not serious, not a self-starter. Social psychologists call this the fundamental attribution error, our habit of blaming people's traits for behavior that was mostly shaped by their situation. You wouldn't call yourself undisciplined for skipping a task that had no scheduled time, no trigger, and no plan for what to do when your kid got sick. But that is almost always what the "homework" you assigned looks like from the inside.
The one question that changes the odds
In the late 1990s, the psychologist Peter Gollwitzer proposed a deceptively small intervention. Instead of asking people to form a goal intention ("I will exercise more"), he asked them to form what he called an implementation intention: a specific if-then plan linking a concrete situation to a concrete action. "If it's Tuesday at 7 a.m. and I've put the kettle on, then I do the ten-minute routine on the kitchen floor."
That's it. No extra motivation, no accountability app, no pep talk. And yet a large meta-analysis Gollwitzer later conducted with Sheeran, pooling nearly a hundred studies, found that implementation intentions had a medium-to-large effect on whether people actually did the thing, across domains from health behaviors to academic work.
The mechanism is worth understanding, because it's what makes the technique feel almost unfair. When you specify the "if," you hand the job of remembering to the environment. The cue (the kettle, the end of the school run, the moment the laptop closes) becomes mentally linked to the behavior, so the situation itself prompts the action instead of requiring the client to summon willpower from nowhere. Gollwitzer called this strategic automaticity: one deliberate decision made in the hot state that then runs on its own in the cold one.
Which means the most useful question you can ask in the last five minutes of a session isn't "Will you do this?" It's "When, exactly, and what will you be doing right before?"
What this sounds like in practice
Watch how differently two endings land.
The first: "Great, so this week try to get the journaling in a few times, and we'll check in next session." The client nods. They mean it. The odds are poor.
The second: "Let's pick one. Just one. When in your week is there already a moment you could attach this to?" The client thinks. "Probably after I drop the kids off. I sit in the car for a minute anyway." "So: if you've just dropped the kids off and you're back in the car, then you write three lines before you start the engine. What's most likely to get in the way?" "Honestly, if I'm running late I'll skip it." "Fine. If you're late, then you write one line instead of three."
That second conversation takes ninety seconds longer. It also does three things the first one doesn't. It shrinks the task to a single behavior, because the research suggests that piling up several plans dilutes each one. It picks a cue the client already encounters, rather than inventing a new habit from nothing. And it pre-decides the failure mode, which the psychologist Gabriele Oettingen's work on mental contrasting suggests is precisely the step that separates optimistic fantasy from plans that survive contact with a real week.
Notice, too, what it does to the relationship. The client stops being someone who owes you a performance and becomes someone you're building a plan with. When they come back, the question is no longer "Did you do it?" with its faint flavor of a parent and a report card. It's "What happened when the cue showed up?" That question gets you information instead of an apology, and information is something you can actually work with.
Your next moves
- Pick one thing, not a plan. Before your next client leaves, choose the single between-session behavior that matters most and let the rest wait. Say the word "one" out loud.
- Ask for the cue, not the commitment. Replace "Will you do this?" with "What will you be doing right before?" Help them find a moment that already exists in their week, and write the if-then sentence down in their words, not yours.
- Pre-decide the failure. Ask "What's most likely to get in the way?" and build a second if-then for it. "If I'm too tired, then I do the two-minute version" is a plan. "I'll try my best" is a hope.
- Change your opening question. Next session, don't start with "Did you do it?" Start with "Tell me about the moment the cue came up." You'll learn whether the cue never fired, the plan was too big, or life simply intervened, and each of those has a different fix.
- Stop grading yourself on their Thursday. Write this somewhere you'll see it: a client's follow-through is a measure of the plan, not of you. Your job is to design a better plan, and now you know how.
Where this meets your calendar
There is a quiet irony here. One between-session commitment nearly all of your clients keep, week after week, without any of this: the appointment itself. It has a when, a where, and a reminder. A booking is an if-then plan with a calendar invite attached. Slate exists to make that part effortless for a solo provider. You run everything from your phone, clients book from a clean web link in a few taps, setup takes about ninety seconds, and it costs half of what Calendly does. That leaves your attention for the harder, more human work: the ninety seconds at the end of a session where you help someone build a plan their Thursday-night self can actually keep. If you'd like the booking side to simply take care of itself, you can start at slate.lumenlabs.works.