How to Track Your Child’s Speech Progress at Home (Without a Clinician)

✓ Reviewed against ASHA Practice Portal, Cochrane systematic review, CDC milestones, ASHA journals (JSLHR/AJSLP)Evidence level BLast reviewed September 9, 2026Published September 9, 2026

Track speech progress at home with two numbers, not impressions: how many of the same 10 words your child says clearly each week, and how much of their everyday speech a stranger understands. Same words, same day, written down. Here is the method, what counts as real change, and when a flat line means see an SLP.

What to measure — and what not to

“Is it getting better?” is the question every parent asks and almost nobody can answer from memory, because a child’s speech varies from day to day and your ear adapts to it. Speech-language pathologists solve this the boring way: they sample the same target under the same conditions and count. You can do the home version of that with two measures.¹

Measure 1: clear words out of ten, for one sound

Choose the one sound you are working on and ten words that contain it — ideally in the position you are practicing (beginning, middle, or end of the word), because a sound can be reliable in one position and absent in another. Once a week, have your child repeat the ten words and count how many were clear. Write the number down. That is the whole measure.¹

Measure 2: how much a stranger understands

Intelligibility — the share of a child’s everyday speech that an unfamiliar listener understands — is one of the most useful single signals, because it integrates everything: sounds, words, and sentences. Measuring it precisely is hard enough that researchers have built whole methods around it, but a parent’s approximation works: ask someone who does not hear your child daily to listen to a minute of ordinary talk and say whether they understood almost none, about half, most, or nearly all.²

📝

What not to track

Don’t track “good day / bad day,” how much your child talks, or how you feel about it. Don’t score conversation in the moment — that turns talking into a test. And don’t compare against other children; compare against last month’s number for your own child.

The weekly 10-word method

1Pick the target sound and its position. Write ten words that contain it. Keep this list fixed for at least six weeks.
2Pick a day and a time — Saturday morning, say — and keep to it. Tired children speak differently from rested ones.
3Sit face to face, somewhere quiet. Say each word once; your child repeats. Do not correct or coach during the check — you are measuring, not teaching.
4Mark each word clear or not clear. When it is not clear, note what you heard (a “th” for S, a W for R, the sound dropped).
5Once a month, ask an unfamiliar listener the intelligibility question and record their answer.
6Put both numbers in a table like the one below. The table is what you show a clinician.

A home speech-progress log. One sound, ten fixed words, one check a week.
Week	Sound · position	Clear / 10	What the misses sounded like	Stranger understood
1	S · initial	3 / 10	“th” on 6, dropped on 1	about half
2	S · initial	4 / 10	“th” on 6	—
3	S · initial	6 / 10	“th” on 4	—
4	S · initial	7 / 10	“th” on 3	most
5	S · initial → final	5 / 10	final S dropped on 4	—
6	S · final	7 / 10	dropped on 3	—

A home speech-progress log. One sound, ten fixed words, one check a week.

Notice week 5 in the example: the number went down because the position changed, not because the child got worse. That is why the sound-and-position column matters — a chart without it will mislead you.

💡

Keep the check separate from practice

Practice happens most days and is where you model, praise, and coach. The check happens once a week and is where you stay quiet and count. Mixing them inflates the score and hides the pattern.

What real change looks like, and how fast it comes

Two signals count. A trend across three or more weekly checks is progress; a single good week is not. And a change in position — a sound that was clear only at the start of words becoming clear at the end — is progress even if the raw number dips while the new position is learned.

3×/week > 1×/weekin a randomized trial of 54 preschoolers, practicing 3×/week for 8 weeks beat the same cumulative dose at 1×/week over 24 weeks — and once-weekly was statistically indistinguishable from a storybook control.⁸

That trial is the honest answer to “how fast.” Speed follows dose: the same total amount of practice, spread thin, produced no measurable advantage over a storybook control. Motor-learning research in speech treatment points the same way — many practice trials, distributed over time, beat one long session. If your weekly number is flat and practice has been once a week, the fix is frequency before anything else.^8,11

Does therapy work at all? The Cochrane review of speech and language therapy for children with primary speech and language delay or disorder found evidence of benefit, particularly for expressive phonology, and found no significant differences between clinician-administered intervention and intervention implemented by trained parents. A systematic review of parent involvement in speech-sound intervention supports parents delivering practice between sessions. Your log is the evidence that the parent-delivered part is happening.^9,10,12

Reading the numbers against age

A 10-word score only means something against what is typical for your child’s age. Consonants arrive over years, not months: in a large review of U.S. studies, 90% of children had acquired /s/ and /z/ between 4;0 and 4;11, with sounds like /r/ and “th” later still. A three-year-old scoring 3 of 10 on S is on schedule; a six-year-old scoring 3 of 10 is not.¹³

For the intelligibility measure, the CDC’s 3-year milestone is that a child talks well enough for others to understand most of the time, and a 2021 normative study found children should be at least 50% intelligible to unfamiliar listeners by their fourth birthday — while clarity keeps growing for years after that, so an imperfect four-year-old is normal. The old rule that a four-year-old should be 100% intelligible traces to a 1988 screening study and is contradicted by the newer growth data.^6,4,3,5

How to read a stranger’s answer by age, using the milestone and growth-study benchmarks cited above.
Age	Benchmark	Act if
By 3	Understood most of the time (CDC 3-year milestone)	A stranger understands little of a 3-year-old’s speech
By 4	At least 50% intelligible to unfamiliar listeners	A stranger understands less than half at 4
5–7	Clarity still improving toward near-complete intelligibility	No upward movement across several months

How to read a stranger’s answer by age, using the milestone and growth-study benchmarks cited above.^6,4

NIDCD’s milestone checklist is a good companion for the language side — words, sentences, following directions — which this page does not measure.⁷

When a plateau means see an SLP

A flat line after four to six weeks of consistent, near-daily practice is a signal, not a failure. It usually means the target is wrong (a sound the child is not developmentally ready for), the technique is wrong (drilling a lateral S reinforces the wrong tongue position), or something else is going on that a home log cannot see. Choosing targets and techniques is the clinician’s job — and your log makes their evaluation faster and better.¹

Waiting is the one strategy the evidence argues against. In a population cohort, 3.6% of 8-year-olds had persistent speech sound disorder, and ASHA’s guidance on early identification is unambiguous that earlier is better.^14,15

🩺

Book an evaluation if

Four to six weeks of consistent practice show no trend; a stranger understands less than half of your 4-year-old; your child is frustrated or avoiding talking; or the misses are wet and slushy rather than “th”-like. Bring the log. ASHA ProFind lists certified speech-language pathologists.

How SpeechStep tracks it — and the free tracker below

Two different things live on this site, and it is worth being precise about them. The free progress tracker below is a manual chart: you type a weekly percentage per sound, it plots six weeks, marks a sound as mastered at 90% or above, and lets you share the record with a therapist by link or email. It starts with editable sample rows so the chart is not blank — overwrite them with your own numbers.

Signed-in practice in the app is measured, not typed. Each attempt at a word gets a clarity score and a coaching tip, and the progress page shows attempts per sound, weekly clarity from real-scorer attempts, and the streak — every number comes from stored attempts, and gaps are shown as gaps rather than filled in. It does not diagnose anything, and it is not a substitute for the evaluation described above; it is the daily half of the same log.¹⁶

Start the chart with this week’s 10-word score:

Free Speech Progress Tracker

Enter a weekly percentage per sound and chart six weeks — then share the record with a therapist.

Try it free →

Frequently asked questions

What is the simplest way to track my child’s speech progress?＋

Pick the one sound you are working on and ten words that contain it. Once a week, at the same time of day, have your child repeat the ten words and write down how many were clear. Ten weeks of “clear out of 10” tells you more than any general impression, and it is exactly the kind of record a speech-language pathologist can use.

How do I measure whether strangers can understand my child?＋

Ask someone who does not hear your child every day — a relative, a neighbor — to listen to a minute of ordinary talk and estimate how much they understood: almost none, about half, most, or nearly all. Intelligibility to unfamiliar listeners is a recognized measure in speech research, and the CDC’s 3-year milestone is that a child talks well enough for others to understand most of the time.

How fast should progress happen?＋

It depends on the dose. In a randomized trial of 54 preschoolers, children practicing 3 times a week for 8 weeks improved more than children given the same total amount 1 time a week over 24 weeks. With short daily practice, a weekly 10-word check often moves within a few weeks; with once-a-week practice it may not move at all.

What counts as real change on my chart?＋

A single good week is noise — children have clear days and mumbly days. Look for two things: a rising trend over three or more weekly checks, and a change in position (a sound that was only clear at the start of words becoming clear at the end). One week up, one week down, is not yet progress.

When does a flat line mean I should see an SLP?＋

If four to six weeks of consistent, near-daily practice show no change on the 10-word check, stop practicing harder and book an evaluation. A plateau usually means the target or the technique is wrong, not that your child is not trying — and choosing targets is a clinician’s job.

Does SpeechStep track progress automatically?＋

For signed-in practice, yes: every attempt gets a clarity score and a coaching tip, and the progress page shows attempts per sound, weekly clarity from real-scorer attempts, and the streak — nothing is estimated or filled in. The free progress tracker on this page is separate: you type in a weekly percentage per sound yourself and it charts six weeks.