How to Use Data to Plan Therapy in Minutes a Week

If you’ve ever sat down on a Sunday night to “plan therapy” and had no idea where to start, here’s the good news: with the right structure, you shouldn’t need to.

Most of the planning most SLPs are doing happens outside the session, at a desk, disconnected from what the student actually did last time. But when your session has a consistent structure and your data feeds directly into it, most of your planning happens in the session. The data tells you what to do next while you’re still sitting across from the student, and by the time you wrap up, you already know what next session looks like.

That’s what this post walks through: a structure that turns your data into decisions in real time, instead of a separate planning task you have to find time for later.

Why Structure Comes First

Data only drives planning if you have somewhere for it to plug into. Without a consistent session structure, data is just a number you write down. With one, it becomes the input for a real-time decision.

Two things need to be in place before the data-based part of this works:

A consistent session structure. 81% of SLPs report they don’t use a structured framework, or use one inconsistently, even though the research supports it. A predictable structure reduces cognitive load for students, supports self-regulation, and gives you a real-time clinical decision tree instead of a guessing game. We break this down in full in Session Structure for School-Based SLPs , so this post assumes that foundation and focuses on where data plugs in.

Thematic units as your context. Planning gets a lot easier when you’re not reinventing materials for every goal, every session. A thematic unit gives you a consistent context to hang your Teach and Practice phases on, so you’re adapting one plan across students and goals instead of building from scratch each time. More on that in Thematic Units for Therapy Planning (this month’s podcast episodes go deeper on this too).

With structure and thematic units in place, data has something to plug into. That’s what the rest of this post is about.

The Session Structure, and Where Data Drives It

A structured session has five phases: Check In, Assess, Teach, Practice, and Wrap Up. It doesn’t have to run in a strict straight line: some sessions lean heavy on Teach, some barely need it, and you can loop back a step if a student gets dysregulated mid-session. Data does the most work in the middle three phases.

1. Check In. Get the student settled and oriented before you dive in. This is a good spot for a quick temperature check (how are they arriving today?) and for reviewing goal cards, a simple way to keep students aware of what they’re working on and why it matters to them. It takes a minute and pays off in engagement.

2. Assess (Probe). Run a quick probe using materials or a context you’re not actively treating with, and only praise participation, not performance, so you get a clean read on where the student actually is. For a small group, this can take about three minutes total, one quick probe per student, before you move into the rest of the session. This single data point is what drives everything that happens next.

3. Teach or Practice, based on the score. This is the decision point. Your probe score tells you exactly where to spend your time (these are rough ranges to guide clinical judgment, not hard cutoffs):

• 0 to 30% accuracy: Teach. The student needs direct instruction before practice will be productive. Teaching first, before jumping to practice, improves understanding and performance (Swanson, 2001). Skip this and a student at 0% will just get frustrated trying to guess what you want.
• 30 to 70% accuracy: Review the teaching, then practice with support. The student has partial understanding; reinforce it, then give them a chance to apply it with scaffolding.
• 70 to 100% accuracy: Practice. The student’s ready to apply the skill in a more authentic context, with feedback rather than instruction. Defaulting to teaching here wastes session time the student doesn’t need spent that way, and it can hurt engagement too.

This is the core of data-based planning: you’re not deciding what to do based on what you planned last week. You’re deciding based on what the student showed you five minutes ago.

Having your visuals ready ahead of time makes the Teach phase run smoothly. A simple approach: one placemat-style visual per student, tied to the one goal they’re focusing on that session, so you’re not digging for materials mid-session.

4. Practice. Use authentic contexts (Gillam et al., 2014), curriculum-based or literacy-based therapy, and real feedback (Proctor-Williams, 2009). Or hand it to the student: let them self-correct, and use your prompting hierarchy to provide only as much support as they need, avoiding negative practice.

5. Wrap Up. Recap the session together: what went well, what didn’t, and why. Make a plan for next time, including anything the student needs to do at home, and make connections, helping the student see how the skill applies outside the session (sharing a read-aloud link for this month’s book is an easy one).

Capture the Plan Before You Walk Out

The Wrap Up phase is where most of your “planning” actually happens, and it takes minutes, not a separate prep session. The best therapy planning happens right at the end of a session, while everything (how the student responded, which strategies actually helped) is still fresh. Waiting until later means you’re stuck trying to reconstruct all of that from memory.

Once you’ve logged the session’s data (probe score and level of support), generate your session note. If you build a spot into your note template for a quick therapy plan, jot down what you want to remember for next time before you move on to the next student. That one line (“still at 30% on initial sounds, needs one more teach cycle” or “hit 80%, ready to shift to practice with peers”) is often all the “planning” a session needs, and it’s a gift to your future self who won’t have to reconstruct it from scratch.

Let Level of Support Round Out the Picture

Probe accuracy tells you what phase to start in. Level of support tells you how the student is actually progressing within and across sessions, and it’s often the more useful signal for deciding whether to hold steady or move on.

A student can plateau at 65% accuracy for a few weeks and still be making real progress if the support they need has dropped from maximal cues to an occasional verbal reminder. Track both: independent, minimal verbal cue, moderate visual cue, maximal support. If you want a ready-made framework, I put together a Level of Support Rubric that walks through the levels and how to use them consistently.

Start With Clear Goals

None of this works without goals that are specific enough to probe against. Before you can run a meaningful probe, you need: a present-levels picture of the student, a specific goal, a baseline, and a way to keep measuring progress.

SLP Now’s assessments handle this pipeline directly. You run a present-levels assessment (broken into small, specific sections so you’re only running what you actually need), and it surfaces goal recommendations based on the student’s responses. From there, you collect a baseline probe for each recommended goal right in the assessment. Anything the student’s already strong on, skip; anything they’re not, that becomes a goal, automatically populated with its own probe, visuals, evidence-based treatment strategies, and therapy activities attached. The point isn’t the software specifically; it’s that clear goals with a probe already attached are what make the Assess phase fast instead of something you have to build from scratch every time an IEP is due.

Review the Trend, Not Just the Session

Within a session, your probe score drives the decision. Across sessions, it’s the trend that matters.

When you’re reviewing a student’s data, pull up their graph and look for the pattern: Is accuracy climbing? Flat? Is the level of support they need dropping even if accuracy isn’t? A graph makes that obvious in a few seconds in a way that flipping through session notes doesn’t. If the trend is flat or declining across several sessions, that’s your cue to change something: the teach approach, the materials, the context.

The Weekly Check-In

Because most of the planning happens in the session itself, your weekly review can be short. That’s the real payoff of this whole system: when your session structure, thematic units, and goal-specific tools are set up ahead of time, your actual planning takes minutes a week, not hours.

Set aside a little time (Friday afternoon works well for a lot of SLPs) to skim your notes from the week, prep anything you’ll need for the week ahead, and get your head into the right space for Monday. You’re not building session plans from scratch here; you’re just making sure nothing slipped through and you’re walking in prepared.

What About Mixed Groups?

The structure above works with groups, not just one-on-one, since a probe, a teach phase, and a practice phase are all things you can run across multiple students in the same session slot. If you want strategies specifically for managing multiple students with different goals at once, How SLPs Take Data in Mixed Groups covers that in depth.

Going Digital

This entire structure gets easier to run consistently when your tools are connected instead of scattered across a binder, a spreadsheet, and a stack of sticky notes.

SLP Now ties your assessments, goals, probes, and session notes together: present levels and baseline data auto-populate a goal with its probe, visuals, and evidence-based strategies attached, session notes generate directly from your logged data, and progress graphs update automatically so the trend is always one click away. Want to set it up? The free Digital Data Bootcamp walks you through it step by step.

Common Questions About Data-Based Therapy Planning

How do I use probe data to decide what to do in a session?
A quick probe at the start of the session tells you where to focus. Scores of 0 to 30% mean the student needs direct teaching before practice will help. Scores of 30 to 70% mean reviewing the teaching, then practicing with support. Scores of 70 to 100% mean the student is ready to practice in a more independent, authentic context.

Do I need to plan sessions in advance if I’m using data this way?
Not in the traditional sense. With a consistent session structure and thematic units in place, the probe at the start of each session tells you what to do that day. Your real planning work is setting up the structure, the thematic unit, and the goal-specific tools ahead of time, not scripting each individual session.

What’s the difference between accuracy and level of support, and why track both?
Accuracy tells you how often a student succeeds during a probe. Level of support tells you what it took for them to get there. A student can plateau on accuracy while steadily needing less support, which is real progress you’d miss if you only tracked the percentage.

How often should I review data outside of sessions?
A short weekly check-in is usually enough, since most of the decision-making already happens during the session itself via the probe. Use that time to skim your notes, prep materials for the week, and catch anything that needs a closer look.

Does this approach work for mixed groups?
Yes. A probe, teach phase, and practice phase can all run across multiple students with different goals in the same session. See How SLPs Take Data in Mixed Groups for group-specific strategies.

---

Ready to build a system where your data does the planning for you? Start with the free Digital Data Bootcamp to get your system set up step by step. Or try SLP Now free for 14 days and see how the assessment-to-goal-to-session pipeline works in practice.

x Marisha