Thursday evening, you're looking over tomorrow's schedule. A client you've seen for six weeks is on it. Last session went well; you built a plan together, she nodded, and wrote it down. Now you have a quiet suspicion she'll arrive with little to report, or send a cancellation text in the morning.
More often than not, these types of situations come down to the plan made between sessions, not necessarily the specific client.. So how do you keep therapy clients engaged between sessions? The simple answer: build the time between sessions into the treatment. Tasks that fit the client's life and current abilities, practiced in session before they're tried alone, and asked about at the next session. Apps and reminders can support that plan, but they can't replace it.
This guide is for solo practitioners who want a repeatable system, early-career clinicians still building instincts, and experienced therapists who suspect their between-session work could be stronger. We'll go past "assign homework" and into the situations that actually stall engagement: ADHD, OCD, trauma histories, clients with limited access to technology, and clients you see only once a month.
Key Takeaways:
- Treat the week between sessions as part of the therapy. Plan it, rehearse it, and review it the way you would any in-session intervention.
- A strong between-session task has five parts: a clear purpose, a specific cue, a fit with the client's current abilities, a rehearsal in session, and a review next time. This checklist is our synthesis of the research and clinical reasoning, not a validated protocol.
- "Homework" is a loaded word for some clients. Alternatives like "action plan" or "experiment" can lower the pressure.
- A missed task is information, not a verdict. It often points to a mismatch between the task, the timing, and the client's current capacity.
- ADHD, OCD, and trauma histories each change what "doable" looks like. Standard homework needs adapting for all three.
- Digital tools work best with a human attached. Pick one tool, set it up together, and review what the client did with it.
- Know where the evidence is strong and where it isn't. Evidence for between-session work improving outcomes is stronger than the evidence for it boosting retention.
Why Between-Session Engagement Matters
Work done between sessions is a relatively well-supported idea in psychotherapy, but it comes with an important asterisk: most of that support is within a CBT framework. A 2024 review of between-session homework found substantial evidence within cognitive-behavioral therapies and called it a promising method across approaches, while noting that research beyond CBT is still, unfortunately, limited. For OCD, a systematic review of ERP homework found that completing between-session assignments robustly predicts outcomes in both adults and youth.
Research is also starting to suggest that the quality of the tasks in between sessions is more important than the amount of homework given or practiced. A meta-analysis of homework compliance in CBT was able to link both quality and quantity to outcomes, but evidence suggested the quality of the work was more correlated with symptom change in some contexts.
In practice, this looks like changing the approach to assigning homework, and reframing the questions you ask at the start of the session. Instead of asking, "Did you do the homework/task/challenge?" you might want to consider asking, "What happened when you tried?" or “what made it difficult to do?” These questions could tell you more information about how the task fit, where it broke down, and what you and your client might want to adjust. If you only check the amount of times a client completed the homework, you might miss out on the most useful information. .
What about retention? In a study of 176 adults, specifically in cognitive therapy for depression, clients whose therapists emphasized homework and behavioral methods early on were less likely to drop out. In a small trial of cognitive processing therapy for PTSD, most clients who never turned in the early written assignments ended up leaving treatment.
Neither study shows that improving your homework practices will guarantee a reduction in dropout. But they point to something clinically useful: an early skipped task may be one of the first visible signs that a client is wavering. Prioritizing a curious conversation early on about homework costs far less than finding out your client has been avoiding assignments and feeling uncomfortable coming to session.
The Five Elements of a Strong Between-Session Plan
Think of this as the checklist to run through before the client leaves the room. Each element comes with sample language you can adapt to your own voice.
1. A Clear Purpose
The task should connect to something the client wants, their treatment goals, and to your working formulation. A task that is sound in technique and modality can still prove to be ineffective if it’s not matched well to the client, or their needs. A client who is naturally reflective might be more amenable to journaling about their experiences or monitoring them in worksheets, while a client who is more action-oriented might do better with experiments and real-life practice. A client working on their anxiety might not feel that a worksheet on self-compassion is relevant to their goals, even if it is what the modality recommends.
Sample language: "Last week you said mornings are the hardest part of the day for you. What do you think would help you get through them this week? I have a couple of ideas, but I'd like to hear yours first."
2. Specificity: A Cue, a Place, and a Duration
Vague tasks often fail quietly. Without specifics, it’s easy to fall off track. Specifying how, when, where, and for how long can improve follow-through.
Instead of: "Practice your breathing exercise this week."
Try: "After you pour your coffee tomorrow morning, do the three-minute breathing exercise at the kitchen counter. Do it again after dinner if you can."
3. Fit with Current Capacity
Some tasks assume a skill the client hasn't built yet, like noticing a thought as it happens or staying with distress. When that's the case, the problem isn't effort, it's fit. The fix is to scaffold: break the task into smaller steps. Things they could do even on their worst day. Then, practice it together, and add complexity gradually as the client shows they can handle it.
Sample language: "Before you try this on your own, let's do one together right now. If it feels too hard, we can adjust and make it more manageable."
4. Rehearsal in Session
When clients practice a task in session and see that they can do it, they may feel more confident trying it at home. Even a couple of minutes of rehearsal in the room can make the task feel less foreign, and can improve client follow-through.
Sample language: "Let's log your mood together right now so you know exactly what it looks and feels like. Then you'll repeat it at home, just like we practiced together."
5. Review at the Next Session, Every Time
If you don't ask about the assignment, clients might lose the accountability piece of it. Forgetting, dismissing, or avoiding the conversation has the potential to model to your clients that the assignments don’t actually matter. Open with a brief question about the experience, not only completion.
Sample language: "Let's start with how the plan went. What did you notice? What got in the way? What would you change?"
The language matters too. A recent paper in the Journal of Clinical Psychology notes that "action plan" has been proposed as an alternative to "homework," and that the school-like framing can stir up anxiety or shame in clients sensitive to evaluation. Other clinicians favor "experiment" or "something to try." Try out different wording with your clients and see what fits.
Scenario-Based Guidance: When the Standard Approach Falls Short
Scenario: The Client with ADHD
For adults with ADHD, follow-through is a core challenge, not a motivation problem. Consider reviewing each person's assignment at the start of every session, preview the next one at the end with some advanced troubleshooting, and remind clients that not being able to complete the exercise will never be judged or shamed in session. The idea here is to keep clients from skipping sessions when they didn’t weren’t able to complete the exercise, or had trouble even trying.
Ideas that follow from the same logic:
- Body Doubling: Start the task in the room with the client. For someone with ADHD, starting a task is much easier with someone next to you.
- Use external cues: Reminder notes in places the client passes often, and phone alarms with labels that say exactly what to do, can help keep the task top of mind.
- Shrink the plan: One thing that actually gets done beats five things that have just been planned, but not executed on.
Sample language: "You said journaling every day isn't realistic. What if it's one sentence, on the sticky note by the coffee maker? How would that feel to do? More or less manageable?"
Scenario: The Client with OCD
For OCD, between-session practice is a large part of the treatment. A vague review question like "How did exposure go?" can miss a lot. Ask what exactly the client tried, how long they stayed with it, their levels of discomfort before and after, and what happened next.. If their practice is slipping, look at the assignment’s specifics first. It may be too difficult, too vague, or not clearly tied to what matters to the client.
Sample language: "Walk me through the exposure from start to finish. How long did you stay with it, and what did you do when your urges showed up?"
Scenario: The Client Who Hasn’t Tried
Resist the urge to reassign the same task to the same client when it’s clear they are not willing or able to try it. Usually, this situation signals a mismatch between the task, timing, or the scaffolding. As mentioned before, approaching the situation with validation and curiosity will yield much better alliance and rapport than pushing the client to try again without inquiring about their internal process.
Sample language: "I'm glad you told me you didn’t do the assignment. Let's figure out what we can learn from this week. What was it about the task that made you not want to engage??"
Scenario: The Perfectionistic or Ambivalent Client
For clients who treat every task as a test, frame the assignment as an experiment where any and all results are useful. How a task goes, including what got in the way, is valuable clinical material, and perfection isn't the goal. For ambivalent clients, shift from assigning to inviting: offer two or three options and let them choose what they think fits best for them.
Sample language: "There's no wrong answer. If it doesn't go as planned, you've learned something about what works for you, and we can always adjust as needed."
How to Document Between-Session Plans in Your Notes
A plan you can't see in the record is hard to follow up on, and could prove to be a liability. Brief, specific notes can help you keep track, and make next week's review easier.
Sample language (illustrative; adapt to your documentation requirements): "Between-session plan developed collaboratively: client will [task] at [cue/time/place] for [duration]. Task rehearsed in session; client able to demonstrate. Anticipated barrier: [barrier]. Client's plan if barrier arises: [plan]. To be reviewed at next session."
Sample language for a follow-up note: "Plan from [date] reviewed. Client completed [portion/none]. Barriers identified: [barriers]. Task adjusted as follows: [changes]. Client's response to adjustment: [response]."
Common Mistakes That Could Undermine Engagement
- Assigning without a purpose the client can state. If they can't explain why they're doing it, they’re not likely to do it.
- Vague tasks. "Work on your thought records" has no cue, place, or duration.
- Too many tasks. Assigning multiple tasks at once can be overwhelming and frustrating to clients.
- Never reviewing. Without a review, tasks stop mattering.
- Treating a miss as a failure. It teaches clients to hide what happened.
- Reassigning the same task unchanged. If it didn't work once, ask why before repeating it.
- Piling on tech or apps. More tools could mean more to abandon.
- Assuming access. Ask about devices, data, and Wi-Fi early.
- Copying last week's plan forward. Update it to reflect the current session.
Follow-Up: Tracking Engagement Across Treatment
A single good plan helps. A pattern across sessions tells you more.
- Open each session with the plan from last time. Make it the first agenda item.
- Watch for early drift. In the first few weeks, a skipped task might be worth a closer look.
- Adjust based on what you learn. Smaller, differently timed, or differently framed tasks are all legitimate responses.
- Update the treatment plan when you change your approach in a meaningful way, and note why.
For Solo and Small Practices: Build a Simple System
If you don't have a team or an institutional protocol, a lightweight routine does a lot of the work:
- Reserve the last five minutes of every session for the plan.
- Keep a one-page template with the five elements, so the plan is the same shape every time.
- Choose one tool or format you recommend to most clients, and know how to set it up.
- Add a standing agenda item to check-in at the start of each session.
A Final Word
If there's one idea to take from this guide, it's that engagement between sessions is something you build, not something clients either have or lack. Protect the last five minutes of the session. Agree on one small, specific task, rehearse a piece of it, name the obstacle before it arrives, and ask about it first thing next session.
Frequently Asked Questions
How do you keep therapy clients engaged between sessions?
Treat the time between sessions as part of the treatment. Design tasks collaboratively, make them specific and matched to the client's current capacity, rehearse them in session, anticipate obstacles, and review them at the next session.
What should I do when a client doesn't complete homework?
Explore it with curiosity before changing anything. Ask what got in the way, whether the task was too hard or poorly matched, whether the purpose was clear, and whether the timing worked. Then adjust, scaffold, or rehearse.
Does homework reduce dropout?
The evidence is limited. Studies show associations: early attention to homework predicted lower dropout in one depression study, and skipped assignments flagged dropout risk in a PTSD trial. Neither proves that homework prevents dropout.
Should I call it "homework"?
Not necessarily. For clients who are sensitive to evaluation, "action plan" or "experiment" may feel less loaded. It’s worth it to talk to your clients directly about their preference.
Are apps worth recommending?
They can be, when tied to a task you've already built and paired with human support. Choose one tool, set it up together, review what the client did, and make sure a non-digital version is available.
How should I adjust for clients I see biweekly or monthly?
Consider using smaller, more specific tasks, a written summary, and, where possible, a brief midpoint check-in.
