IEP goals · Special education · Progress monitoring

How to Write Measurable IEP Goals: The Formula, Six Rewrites, and a Checklist

A goal is measurable when two people reading it cold would agree, at the annual review, whether it was met. Here is how to write one every time.

By Shubham Bhatt, Founder, GetWorksheetUpdated October 17, 2026 5 min read

What the law asks for

IDEA requires every IEP to contain measurable annual goals, plus a description of how progress will be measured and when reports go home (34 CFR 300.320(a)(2) and (a)(3)). It does not prescribe the wording. The practical test: could two people, reading the goal with no other context, agree at the end of the year on whether it was met? If the answer depends on anyone's judgement, the goal is not measurable yet.

The six-part formula

A measurable goal answers six questions in one sentence:

  1. By when. The annual review date, or "by the end of the IEP period".
  2. Under what condition. Given a grade-level passage, given a 10-item worksheet, during unstructured recess.
  3. Who. The student, by name.
  4. Will do what. An observable verb: read aloud, solve, write, raise a hand. Not "understand" or "improve".
  5. How well. The criterion: 90 words correct per minute, 8 of 10 items, fewer than 2 prompts.
  6. Measured how, and how often. One-minute timed readings weekly, teacher-made probes every two weeks, a frequency count each day.
By May 2027, given an unpracticed second-grade passage, Maya will read aloud at 90 words correct per minute with at least 97% accuracy on 3 of 4 consecutive weekly probes, as measured by one-minute timed readings.

The condition fixes the difficulty, the criterion is a number the probe produces, and "3 of 4 consecutive" rules out a lucky day.

Six vague goals, rewritten

Before and after, using the formula.

Reading comprehension, grade 4

Before: Student will improve reading comprehension skills.

After: By the annual review, after reading a fourth-grade passage of 200 to 300 words, Jordan will answer 8 of 10 literal and inferential comprehension questions correctly on 3 consecutive biweekly probes.

Write this for my student Make the probes

Math, grade 3

Before: Student will know multiplication facts.

After: By the annual review, given a two-minute probe of mixed multiplication facts 0 to 10, Amir will write 40 correct digits with no more than 2 errors on 3 of 4 consecutive weekly probes.

Write this for my student Make the probes

Writing, grade 2

Before: Student will write better sentences.

After: By the annual review, given a picture prompt and 10 minutes, Sofia will write 4 complete sentences that each begin with a capital letter and end with punctuation, on 4 of 5 weekly samples scored with a checklist.

Write this for my student Make the probes

Behavior, grade 1

Before: Student will not call out in class.

After: By the annual review, during 30-minute whole-group lessons, Eli will raise his hand and wait to be called on before speaking in at least 80% of opportunities, with no more than 2 call-outs per lesson, on 4 of 5 consecutive daily frequency counts.

Write this for my student Make the probes

Social skills, kindergarten

Before: Student will make friends.

After: By the annual review, during 20-minute free play with one peer, Priya will initiate a shared activity using words or a picture card at least 3 times per session, on 4 of 5 consecutive weekly observations.

Write this for my student Make the probes

Life skills, high school

Before: Student will become more independent.

After: By the annual review, given a written shopping list of 5 items and a classroom store, Marcus will locate all 5 items and pay with the next-dollar strategy with no more than 1 verbal prompt, on 3 consecutive weekly trials.

Write this for my student Make the probes

Criterion traps

  • A percentage with no count. "80% accuracy" on a 3-item probe is 2.4 items. Write the count: 8 of 10.
  • "4 out of 5 trials" with no definition of a trial. Is a trial one item, one probe, or one week? Say which.
  • Single-shot mastery. One good day is not mastery. Require consecutive probes, or 3 of 4.
  • Goals written as an absence. "Will not hit" cannot be counted. Write the replacement behavior and its rate.
  • Goals that measure the adult. "Will be given visual supports" is a service, not a goal.

The checklist before you file it

  1. The verb is something you could video.
  2. The condition names the material and its level.
  3. The criterion is a number the probe produces, with the count behind any percentage.
  4. Mastery needs more than one data point.
  5. The measurement tool and its frequency are named.
  6. The baseline in the present levels uses the same measure as the goal.
  7. You already know which probe you will use next week.

The goal is only as good as its probe

"Decode CVC words" needs parallel 10-word lists. "Read 90 words correct per minute" needs timed passages at the same level every week. "Raise a hand to speak" needs a frequency count in a defined period. Decide the probe while you write the goal and put it on the calendar; a goal with no scheduled measurement is a wish. The probe generator builds parallel forms and a record sheet from the goal sentence; the reading fluency guide shows the full cycle for one skill.

This article is drafting guidance for teachers, not legal advice. Your state and district forms govern the final wording.

Free tool

Type the present levels and the skill; get goals written in this formula, ready to edit.

Open the IEP goal generator

Free to try, no card. Results save to your library.

Good to know

Questions, answered

As many as the present levels justify and the team can measure, which in practice is usually two to five. Every goal needs its own probe and its own data; ten goals with no data are worse than three with weekly numbers.

Short-term objectives break a goal into component skills (first CVC words, then blends). Benchmarks mark the expected level at points in the year (60 WCPM by December, 75 by March). IDEA requires them only for students on alternate assessments, but many states and districts still use them.

Use whatever the probe actually produces. Accuracy tasks (decoding, computation) take a count and a percentage: 8 of 10 items. Fluency tasks take a rate: words correct per minute. Behavior takes a frequency or duration in a defined period. Never write a percentage without the count behind it.

Yes. If the data show the goal was met early or is out of reach, the team reconvenes or, where the parent agrees, amends the IEP without a meeting. A goal that the graph shows is wrong should be changed, not waited out.