Twelve goals I actually write, with the practice and the data sheet that go with each. Copy the wording, or hit "Build this goal" to fill in your student and criterion.
Every goal follows the same shape: condition, behavior, criterion, how it's measured. Criteria are starting points. For a student at 20% today, 80% across three sessions is a year's work; for a student at 60%, it might be a semester. Data sheet numbers refer to the free starter pack.
One-to-one counting to 10
Math
Goal wording
By [date], when given a set of up to 10 objects or pictures and asked "How many?", [Student] will count each item once and state the total, with 80% accuracy across 3 consecutive sessions, as measured by trial-by-trial data.
How I teach it: start with objects the student can slide to the side as they count, so "counted" and "not counted" are physically separate. Move to touching pictures only once sliding is solid. The most common error is counting past the last object, so I model a hard stop with a big "ten!" and hands up.
By [date], when shown a numeral from 1 to 20 and asked "What number?", [Student] will name it (verbally, with AAC, or by selecting from a field of 3) with 80% accuracy across 3 consecutive sessions, as measured by a numeral probe.
How I teach it: probe in random order, not 1 to 20. Teens are the wall for most of my kids, so I teach 11 to 19 as a separate set and keep 10 and 20 as anchors. Receptive first (point to the 14) if expressive isn't there yet.
By [date], when given an addition problem with sums to 10 and picture supports, [Student] will select the correct sum from a field of 3 with 80% accuracy across 3 consecutive sessions, as measured by trial data.
How I teach it: count all before count on. Every problem shows both groups as pictures the student can touch and count. I don't take the pictures away until the student is at criterion with them. Then the goal gets rewritten "without picture supports."
By [date], when given a subtraction problem within 10 with picture supports, [Student] will cross out or cover the items taken away and select the correct difference from a field of 3 with 80% accuracy across 3 consecutive sessions, as measured by trial data.
How I teach it: "take away" needs a physical action. Crossing out or covering with a hand is the whole lesson for the first few weeks. The language matters: I say "take away" every time until it's solid, then introduce "minus."
By [date], when shown two sets of up to 10 objects or pictures and asked "Which has more?" (or "fewer"), [Student] will select the correct set with 80% accuracy across 3 consecutive sessions, as measured by trial data.
How I teach it: start with big differences (2 vs 9) and line the sets up so one is clearly longer. "Fewer" is a separate goal in practice; most students get "more" months before "fewer." I write them as two objectives under one goal.
By [date], when shown a letter in random order and asked "What letter?", [Student] will name 20 of 26 uppercase and 20 of 26 lowercase letters (verbally, with AAC, or by selecting from a field of 3) across 3 consecutive sessions, as measured by a letter probe.
How I teach it: the letters in the student's own name first, then the high-frequency ones (a, t, s, m). Teach uppercase and lowercase as one card each with both forms until the student can match them, then split the probe. The ABC song is not letter ID.
By [date], when shown a consonant or short vowel and asked "What sound?", [Student] will produce or select (from a field of 3 pictures that begin with that sound) the correct sound for 20 of 26 letters across 3 consecutive sessions, as measured by a letter sound probe.
How I teach it: for non-speaking students, the goal is "which picture starts with this sound," not producing the sound. That's a real reading skill and it's measurable. I pair every letter with one anchor picture and never change it.
By [date], when shown a target word from a list of 20 (functional and high-frequency words), [Student] will read it or match it to the correct picture from a field of 3 with 80% accuracy across 3 consecutive sessions, as measured by a sight word probe.
How I teach it: functional words first (stop, exit, go, eat, mom) because they show up in the student's day. Picture-matching is the on-ramp; the goal gets rewritten "without pictures" once the student is at criterion.
By [date], after a short story or picture is read aloud, [Student] will answer who, what, and where questions by selecting the correct picture from a field of 3 with 80% accuracy across 3 consecutive sessions, as measured by a WH question probe.
How I teach it: one question type at a time. "Who" is almost always easiest because the answer is a person. "Where" is hardest because places are abstract. I keep the story to two sentences and the picture choices to one person, one thing, one place so the question type itself is the cue.
By [date], when shown a picture and a sentence frame ("I see the ___"), [Student] will complete the sentence by selecting the correct word from a field of 3 and reading or voicing the full sentence (speech or AAC) with 80% accuracy across 3 consecutive sessions, as measured by trial data.
How I teach it: one frame for weeks. "I see the" works because it's true of any picture. The student taps the word, then we read the whole sentence together pointing to each word. For AAC users, the frame lives on their device as one button.
By [date], when given a 3-step picture sequence for a daily routine (hand washing, arrival, snack), [Student] will complete each step in order with no more than one gestural prompt per step in 4 of 5 opportunities across 2 weeks, as measured by a task analysis data sheet.
How I teach it: the visual stays in the place the routine happens, at eye level. I prompt from behind so the student looks at the picture, not at me. "First, next, last" language is the same every time. Fade from pointing to the picture, to pointing at the whole strip, to nothing.
By [date], when shown a picture of a person showing a feeling (happy, sad, mad, scared) and asked "How do they feel? What can they do?", [Student] will identify the feeling and select an appropriate coping strategy from a field of 3 with 80% accuracy across 3 consecutive sessions, as measured by trial data.
How I teach it: teach the strategies when nobody is upset. Three strategies, always the same three (breathe, ask for a break, squeeze). Then the real goal is using them in the moment, which goes on a separate frequency sheet, because a kid who can pick "take a breath" on a card and a kid who takes one when mad are at very different places.
Paste it in and I'll build what's missing: a practice deck, printable pages, and a data sheet that fits the goal. The most requested ones get made first and go out free to the whole list.
Written by a K-3 SDC teacher in a public school special day class. These are starting points, not legal language: your district's IEP format and your student's present levels come first. Nothing here is medical or legal advice.