95 words correct per minute. Four weeks later, 109 to 111. Accuracy up from 94% to 96-97%. Comprehension up from 68% to 72-74%.
That’s real data, from one 9th grader’s reading intervention. None of it happens by accident — and none of it gets tracked without a goal built to be scored, not just hoped for.
“Will improve reading comprehension” sounds specific. It isn’t. It’s a direction, not a destination — nobody can look at that sentence in May and say whether it happened.
Three Parts, Or It’s Not a Goal
A measurable goal has exactly three parts. Drop one, and the whole thing stops being measurable, no matter how much detail is packed into the sentence.
- Condition — the exact circumstances the skill happens under.
- Behavior — the observable action, stated with a verb two different adults would score the same way.
- Criterion — the number that ends the argument about whether it’s met.
Miss the condition, and nobody knows what setting or materials the goal even applies to. Miss the behavior, and two case managers watching the same student score it two different ways. Miss the criterion, and the goal never actually resolves — it just sits there, technically in progress, forever.
Where the 109 Came From
Here’s what those three parts look like, built from a real baseline instead of a template.
A 9th grader came in at 95 wcpm against a grade-level expectation around 130 — strong vocabulary, choppy phrasing, the kind of profile where automaticity, not comprehension, was the actual bottleneck. Four weeks of paired oral reading and explicit syllable work later: 109 to 111 wcpm, accuracy 94% to 96-97%, word-study accuracy 58% to 88%, comprehension 68% to 72-74%.
Written as a goal, that trial reads: “Given a 9th-grade instructional-level passage, the student will read aloud at 130 words correct per minute with 95% accuracy, across 3 of 4 trials.” Condition: instructional-level passage. Behavior: read aloud. Criterion: 130 wcpm, 95% accuracy, 3 of 4 trials. Every week of progress monitoring either hits that line or it doesn’t — no argument required.
Same discipline works at smaller scale. A different student’s written-expression goal tracked oral-to-written transfer using a sentence-starter and linking-word scaffold — the formative check came back 11 out of 15. Not a vibe. A number you can graph next to last week’s number.
Weak Versus Strong, By Domain
The pattern holds across every domain on a caseload. Same failure, same fix, four ways.
| Domain | Weak | Strong |
|---|---|---|
| Reading | Student will improve reading fluency. | Given a 9th-grade instructional-level passage, the student will read aloud at 130 words correct per minute with 95% accuracy, across 3 of 4 trials. |
| Written expression | Student will write better paragraphs. | Given a graphic organizer and a writing prompt, the student will produce a paragraph containing a topic sentence, two supporting details, and a closing sentence, with no more than 2 mechanical errors, in 4 of 5 writing samples. |
| Behavior / self-regulation | Student will manage frustration appropriately. | When presented with a non-preferred task, the student will use a taught coping strategy (request a break, use a visual scale) instead of leaving the classroom, in 8 of 10 observed opportunities across 4 weeks. |
| Vocational / life skills | Student will develop job skills. | Given a multi-step vocational task and a visual checklist, the student will complete all steps in the correct sequence with no more than 1 verbal prompt, across 4 of 5 sessions. |
The Number Isn’t a Guess
The criterion in that reading goal wasn’t picked because 130 sounded good. It came from baseline data plus a reasonable stretch — the student’s actual 95 wcpm, plus a year’s worth of realistic growth.
Skip the baseline and you get one of two failures. Set the number too low, and the student meets the goal in October having learned nothing new. Set it too high — 250 wcpm for a kid starting at 95 — and it’s not a goal anymore, it’s a target nobody on the team actually believes in.
Pull the baseline first. Set the number second. Working backward from a criterion that sounds good on paper is how goals end up either trivially easy or quietly impossible.
One Goal, One Skill
The last way goals fall apart: stacking three skills into one to save space. “Will read fluently and identify main ideas and use context clues” isn’t one goal. It’s three goals wearing one goal’s clothes.
Run progress monitoring on that, and watch what happens the first time the student misses it. Which part failed — fluency, main idea, context clues? Nobody can say. The data was never separated in the first place.
Write one skill per goal. It takes more lines on the page. It also actually tells you something the next time you sit down with the data.
That’s the whole difference between a goal that sounds good in the meeting and one that produces a 109 four weeks later.