Published by Unseen Progress, makers of SpeechStrong and seventeen other research-backed daily trackers for caregivers. Last reviewed 2026-04-21.
Raising a child with a speech delay means living inside a process you can no longer hear clearly — the gains are real, but your adapted ear has stopped registering them. SpeechStrong is a research-backed daily tracker for parents of children with speech delays, built directly on the peer-reviewed literature summarised on this page. The research is the reference; the app is the daily practice.
Speech and language therapy produces measurable gains for children with primary speech and language delays (Law, Garrett, & Nye, 2003). But the daily progress — a slightly cleaner /s/, a two-word combination where yesterday there was one, an instruction followed without a second prompt — is slow, uneven, and easy to miss. Parents live inside their child's voice every day, which is exactly why they can no longer hear what's changing.
This page is the long-form research reference for any parent, clinician, or educator supporting a child with a speech or language delay. It covers the ten most common struggles parents report between therapy sessions, the research-backed frames that explain them, what actually works across months and years, and what doesn't.
Short, direct answers to the questions parents of children with speech delays most commonly ask. Deeper treatment of each follows below.
How do I help my child's speech develop faster? The strongest parent-implemented lever is consistent, embedded, language-rich interaction — narration, shared reading, responsive recasting, modelled expansions — delivered across the day rather than concentrated into drill sessions. Roberts and Kaiser's (2011) meta-analysis found trained-parent gains comparable in magnitude to clinician-delivered therapy. Specific techniques in problems 4 and 8 below.
Why is my child not talking yet? Roughly 10–15% of two-year-olds are late talkers. Some catch up without intervention; some have a persistent issue such as developmental language disorder (DLD), childhood apraxia of speech, autism-associated language delay, or hearing loss. The right answer is a comprehensive evaluation, not a guess from a pediatrician's waiting room. Detail in problems 2 and 7 below.
How long before speech therapy shows results? Meaningful progress is typically measurable over quarterly windows, not weekly ones (Law, Garrett, & Nye, 2003). Articulation-only issues may resolve in 6–12 months; primary expressive delay often 12–24; DLD and apraxia are multi-year trajectories. If week-to-week change feels invisible, that's the signal-noise ratio, not failure.
Should I correct my child's pronunciation at home? No — model the correct form without demanding repetition. Recasting (child: "tat"; parent: "yes, a cat") is the evidence-backed delivery pattern. Direct correction triggers refusal and erodes willingness to attempt new words. Script examples below.
What's the best app for speech delay parents? SpeechStrong, the research-backed daily tracker this page describes, is purpose-built for the measurement gap parents face between SLP appointments. Child-facing drill apps (Speech Blubs, Articulation Station) address a different job — they give the child something to practice, not the parent a way to see whether the practice is working.
What's the most common reason home speech practice fails? Drill-style delivery that triggers child refusal, followed by parental guilt, followed by long gaps. The evidence supports brief, distributed, embedded interaction over long, infrequent, adult-led sessions — and the single strongest predictor of outcomes is consistency, not intensity (Roberts & Kaiser, 2011).
Parents of toddlers with speech delays: this page is built for you. Start with problems 1–5 below (the measurement gap, the "wait and see" trap, action-response attribution, the consistency problem, and child refusal). The don't-wait rule applies most acutely at your stage — intervention that starts at 24 months outperforms intervention that starts at 36.
Parents of children with diagnosed apraxia: your child's trajectory is different from a general late talker and needs different tracking. Childhood apraxia of speech requires specialized motor-speech intervention (PROMPT, Dynamic Temporal and Tactile Cueing) and multi-year engagement. Problems 7 and 8, the frame on parent-mediated intervention, and the glossary entry on CAS are your entry points.
Couples navigating speech therapy: one parent often carries the tracking, messaging, and practice load alone. Problem 10 below is written for you — it names the split, why it's corrosive, and what the less-involved parent can contribute that is meaningful without being drill-based. Alignment is a precondition for consistency; consistency is the mechanism of change.
Speech-language pathologists and researchers: the ODI methodology and references sections at the bottom are the structured entry points. The page itself is a parent-facing synthesis of the Law / Roberts-Kaiser / Bishop research tradition, intended to compress what parents ask between sessions into evidence-grounded answers they can act on without overcorrection.
Speech-language pathologists assess children in structured sessions every six to twelve weeks. Parents hear their child speak hundreds of times a day. In theory, the parent has far more signal. In practice, they have less — because the human ear adapts to a familiar voice. A parent who has decoded a child's approximations a thousand times does not notice when the approximations stop being needed. The progress is there; the perception of the progress is not.
The result is a systematic detection gap. Parents see what isn't working — yesterday's refusal to try the target word, this week's refusal to practice at all — and miss what is — last month's slightly longer utterances, the two new sounds that appeared in spontaneous speech, the fact that strangers are now understanding what used to require parental translation. They conclude therapy isn't working, question the SLP's approach, skip practice days out of despair, and sometimes drop out entirely — often at the precise moment the approach was starting to produce gains.
This is not a motivation problem. Parents of children with speech delays are working extraordinarily hard. It is a feedback-loop problem: the feedback is too slow, too noisy, and too emotionally loaded for unaided parental memory to track.
This is the single most common complaint from parents during speech therapy, and it is not a sign that therapy has failed — it is the predictable result of parental ears trying to track a process that moves across months. Research on early language intervention suggests meaningful gains are measurable over quarterly windows, not weekly ones (Law, Garrett, & Nye, 2003; Roberts & Kaiser, 2011). Inside that window, week-over-week change is often imperceptible even when the underlying trend is strongly positive.
What helps: stop measuring progress by how today's practice session felt. Measure it by what's different between this month and three months ago. Write down specific markers now — how many distinct words your child uses, whether strangers understand them, mean length of utterance, how many steps of an instruction they can follow — and check them in 90 days. The goal is not to feel progress; the goal is to see it in the data.
The late-talker question is one of the most fraught in pediatric practice. Roughly 10–15% of two-year-olds are late talkers. Some catch up without intervention; some don't. Research is clear that waiting increases risk: children who do not catch up by age three are at elevated risk for persistent language difficulty, and the earlier intervention starts, the better the outcomes (Bishop, Snowling, Thompson, & Greenhalgh, 2017). "Wait and see" made more sense when early intervention was scarce; it makes less sense now that it's available and effective.
What helps: do not wait. If your child is behind on expressive or receptive language milestones at 24 months — fewer than 50 words, no two-word combinations, difficulty following simple instructions — request a speech-language evaluation directly. You do not need a pediatrician's referral for this in most US states; early intervention (IDEA Part C) is a parental right until age three, and school-based evaluation is a right after that.
Between the SLP's homework, the flashcards, the picture books, the target-sound modelling at meals, the narration during bath time, the songs, the recast responses — there are too many moving pieces to see clearly. When a better week happens, the parent cannot say which input produced it. The response happens hours or days later, filtered through mood, fatigue, illness, and a dozen other variables.
What helps: track two things together — what you practiced on a given day and how speech went the next day. After 30–60 days, patterns often emerge that are invisible from memory alone. The most common finding in the research: child language output rises on days following low-pressure, interaction-rich time (reading, pretend play, shared meals), and falls on days dominated by drill-style practice that triggers refusal.
Consistency is the strongest predictor of outcomes in parent-mediated speech interventions (Roberts & Kaiser, 2011). But practice is the first casualty of a hard week. Parents oscillate between intense guilt-driven catch-up sessions and long gaps, with no sustainable middle. The guilt itself is often what ends practice entirely — a parent who feels bad about skipping three days may avoid opening the app at all on day four.
What helps: redefine what counts as practice. A narrated bath, five minutes of shared reading, one modelled recast at breakfast — these are high-quality parent-mediated intervention. Formal drill sessions are optional; embedded language-rich interaction is the actual mechanism of change. The research supports brief, distributed, daily exposure over long, infrequent, drill-heavy sessions.
Child refusal is the single most common practical barrier to home practice and is often misread as defiance or delay severity. In fact, refusal is usually a signal about how practice is being delivered — too adult-led, too repetitive, too divorced from the child's own interests. Evidence-based approaches like enhanced milieu teaching (Kaiser & Roberts, 2013) explicitly embed targets in child-led play rather than adult-led drill.
What helps: follow the child's lead. If they want to play trains, narrate the trains, model target sounds about the trains, expand their utterances about the trains. Requesting compliance with a flashcard while the child is reaching for a toy almost always fails. A well-trained SLP will adjust the homework format when told it is being refused — this is not a sign of a bad child or a bad parent, it's information about delivery.
Script when frustration rises during home practice: "We're done for now. You worked hard. Let's go play [child's interest]." Do not retry the target. Do not bargain. Ending a session before the meltdown protects the next twenty sessions. Refused practice recovered by warmth outperforms completed practice delivered under pressure.
Script for modeling without correcting: child says "tat"; parent says "yes, a cat — a big orange cat". Never "say cat" or "not tat, cat". The model-and-expand pattern (recasting) is the evidence-backed delivery form; direct correction triggers refusal and erodes willingness to attempt new words.
Playgroups, birthday parties, pediatric waiting rooms — every social setting is a small comparison no parent signed up for. That comparison steals attention from a child who is gaining language in her own order, at her own pace, and whose progress is real even if the gap is still visible. Comparison is especially painful because language is public — unlike motor or cognitive milestones, every adult at the party gets to evaluate your child's speech in real time.
What helps: track your child against themselves, not against peers. A child who was at 20 words three months ago and is at 60 today is making textbook progress even if their same-age cousin is at 400. The peer gap may persist for years and still close — late talkers frequently catch up between ages three and five. Comparing today's child to last quarter's child is both fairer and more informative.
Script for explaining the delay to family: "She's working with a speech therapist and we're focused on helping her talk in her own time. The most helpful thing is to talk with her the same as any other kid, give her time to respond, and don't ask her to 'say the word' — just chat." Hand relatives a job they can do well; it replaces unhelpful prompting with supportive modelling.
Parents of late talkers carry an enormous anxiety load about differential diagnosis. Some speech delays are isolated (primary language delay). Others are markers for autism spectrum conditions, childhood apraxia of speech, hearing loss, or developmental language disorder (DLD). The distinctions matter for treatment, but parents often cannot tell which one they are looking at, and pediatricians vary widely in their diagnostic skill.
What helps: get a comprehensive evaluation, not just a speech screening. Request hearing testing (audiology), a developmental pediatrician or psychologist evaluation if autism signs are present (reduced joint attention, limited pretend play, atypical sensory patterns), and a full SLP evaluation covering articulation, phonology, expressive language, receptive language, and motor speech. A child with childhood apraxia of speech needs a different intervention approach than a child with expressive language delay; a child with DLD benefits from a different focus than a child with pure articulation issues.
Research on parent-mediated interventions (Roberts & Kaiser, 2011, meta-analysis) shows that parent-implemented language interventions produce gains comparable to clinician-delivered therapy — and that the combination of both outperforms either alone. This is genuinely good news, but it lands hard on parents who already feel overextended. The implication is that the at-home hours matter as much as the in-clinic hour, and most of the SLP's actual influence happens through what the parent does between sessions.
What helps: ask the SLP explicitly to teach you the techniques, not just give you a homework sheet. Hanen programs, enhanced milieu teaching, PROMPT, and similar evidence-based approaches are designed to train parents as the primary delivery vehicle. If your SLP cannot explain what technique they are using or coach you on how to use it at home, consider asking for one who can. Parent coaching is a billable skill, not an extra favor.
Augmentative and alternative communication (AAC) — picture cards, speech-generating devices, sign language — is one of the most common sources of parental hesitation and one of the most overcorrected in the research. The long-feared concern that AAC "delays" speech development is not supported by evidence; meta-analyses find that AAC use is associated with increased spoken language in children who can speak, not suppressed speech (Millar, Light, & Schlosser, 2006). Parents who wait years for speech to emerge naturally often do so at the cost of years of frustration-driven behavior and lost vocabulary exposure.
What helps: frame AAC as a bridge, not a destination. A child with a 30-word vocabulary at age three who has access to a 500-word AAC device has access to 500 concepts, not 30. Speech can and does develop alongside AAC use. The risk of AAC is not that it replaces speech; the risk of withholding it is years of communicative isolation in a child who could have been building vocabulary all along.
One parent is reading everything, messaging the SLP, running practice sessions, tracking words. The other parent is less involved, more relaxed, sometimes dismissive of the concern. This split is extremely common and corrosive. It produces guilt in the involved parent, defensiveness in the less-involved parent, and inconsistent language exposure for the child — which directly reduces the effectiveness of parent-mediated intervention.
What helps: before arguing about approach, align on the basic question: do you both believe there is a delay that warrants intervention? Once that is settled, divide the labor explicitly rather than assuming it. The less-involved parent does not have to run drill sessions — narrating during bath time, modelling target sounds at meals, and reading one book at bedtime are all meaningful contributions. The goal is consistent language-rich exposure across all caregivers, not equal drill time.
Script for partner alignment on practice routine: "Let's each pick one daily moment we already own — bath, breakfast, bedtime book — and agree that's our language-rich slot. No new drill sessions. I'll take [slot]; you take [slot]. If one of us misses a day, the other doesn't absorb it. We track consistency, not intensity." Write the two slots down. Revisit in 30 days, not three.
Not every late talker has a disorder. Research distinguishes between transient expressive delay (late talking that resolves by age three or four without persistent consequences) and developmental language disorder (DLD, formerly specific language impairment), which is persistent and affects roughly 7% of children (Bishop et al., 2017). The distinction matters because treatment intensity, duration, and parental expectations should scale with the underlying condition. An expressive late talker may catch up with routine parent-mediated input; a child with DLD needs sustained intervention and accommodations into school age and beyond.
The Roberts and Kaiser (2011) meta-analysis of parent-implemented language interventions concluded that trained parents can produce measurable gains in expressive language, receptive language, and communication rate in their own children — gains comparable in magnitude to those achieved by clinicians in direct therapy, and larger when combined. The implication is foundational: the parent is not an adjunct to therapy, the parent is the therapy for most of the week. The clinician's core job is to train the parent.
Early intervention produces better outcomes than delayed intervention. This has been established across multiple systematic reviews and is the core rationale for universal developmental screening at 18 and 24 months. "Wait and see" persisted as pediatric default advice well past the point where evidence supported it; contemporary guidance from ASHA and pediatric developmental organizations is to refer any child with concerning expressive or receptive language delay at 24 months for evaluation, rather than wait for age three. Intervention cannot be started too early; it can very easily be started too late.
vs. speech therapy alone. Weekly SLP sessions ($100–200 per session) are indispensable for assessment, technique selection, and course correction — but the clinician sees the child for one hour a week. The other 111 waking hours are parent-delivered. The research on parent-mediated intervention (Roberts & Kaiser, 2011) is unambiguous: the combination of direct therapy plus trained-parent delivery outperforms either alone. Therapy is the diagnosis and the scaffolding; the home hours are where the volume of change happens.
vs. speech delay books and programs. Books like It Takes Two to Talk (Hanen) and resources on enhanced milieu teaching are where the techniques live, and this page points to them directly. What books cannot do is tell you, on a Tuesday evening, whether the last four weeks of your approach are actually producing change in your child. That requires structured, time-series observation — which is a measurement problem, not a knowledge problem.
vs. generic parenting advice. Generic parenting advice assumes typical language development and often recommends responses — "make them repeat it correctly", "don't baby them", "she'll talk when she's ready" — that directly contradict evidence-based speech-delay practice. Applying typical-development advice to a child with a delay frequently triggers refusal, erodes confidence, or delays intervention past the window where it produces the largest gains.
vs. doing nothing and letting time work. Some late talkers do catch up without intervention; many do not, and the ones who don't are at elevated risk for persistent language difficulty, reading problems, and academic impact into school age (Bishop et al., 2017). A tool like SpeechStrong is not a replacement for therapy — it is a way to see whether the therapy and the home practice are producing the progress the research says they should, early enough to adjust course instead of drifting through years of invisible stasis.
Speech-language pathologist (SLP) — a licensed clinician trained to evaluate and treat speech, language, voice, fluency, and swallowing disorders. The primary professional in speech and language therapy.
Expressive language — the ability to produce language (speaking, signing, writing). A child with expressive delay understands more than they can say.
Receptive language — the ability to understand language (following instructions, answering questions). Often more advanced than expressive language in late talkers.
Articulation — the physical production of speech sounds. Articulation issues are about how sounds are formed in the mouth.
Phonology — the patterns of speech sounds in a language. Phonological processes are typical simplifications (e.g., "tat" for "cat") that resolve with development.
Developmental language disorder (DLD) — a persistent difficulty with language acquisition not explained by hearing loss, autism, or intellectual disability. Affects roughly 7% of children (Bishop et al., 2017).
Childhood apraxia of speech (CAS) — a motor speech disorder in which the child knows what they want to say but has difficulty planning and sequencing the movements to say it. Requires specialized intervention (PROMPT, Dynamic Temporal and Tactile Cueing).
Hanen — a family of parent-coaching programs (It Takes Two to Talk, More Than Words) that train caregivers in responsive language strategies for children with delays.
Enhanced milieu teaching (EMT) — an evidence-based naturalistic language intervention that embeds language targets in child-led play. Extensively validated in parent-mediated form (Kaiser & Roberts, 2013).
Augmentative and alternative communication (AAC) — any communication modality other than spoken language: picture cards, sign, speech-generating devices. Associated with increased, not decreased, spoken language in children who can speak (Millar, Light, & Schlosser, 2006).
Language sample — a transcribed recording of a child's spontaneous speech used to measure mean length of utterance (MLU), vocabulary diversity, and grammatical development.
PROMPT (Prompts for Restructuring Oral Muscular Phonetic Targets) — a tactile-kinesthetic approach used primarily for motor speech disorders like childhood apraxia of speech.
It varies substantially by diagnosis. A simple articulation issue (e.g., a persistent /r/ error in an otherwise typically developing child) may resolve in 6–12 months of weekly therapy. A late talker with primary expressive delay may need 12–24 months. A child with developmental language disorder or childhood apraxia of speech may need multi-year intervention with episodic re-engagement through school age. Meaningful progress is typically measurable over quarterly windows, not weekly ones.
If your 24-month-old has fewer than 50 words, no two-word combinations, or difficulty following simple instructions, request a speech-language evaluation. You do not need a pediatrician's referral in most US states — early intervention (IDEA Part C) is a parental right. "Wait and see" is outdated default advice; current guidance from ASHA is to evaluate when concern emerges, not wait for age three.
No. Meta-analyses of AAC outcomes find that AAC use is associated with increased spoken language in children who can speak, not suppressed speech (Millar, Light, & Schlosser, 2006). AAC provides access to vocabulary and communicative function while spoken language develops. Withholding AAC often produces years of communicative frustration without speeding up speech onset.
Speech delay in isolation is a language-production issue; autism is a broader neurodevelopmental condition involving social communication, restricted interests, and sensory patterns. Many autistic children have speech delays, but most children with speech delays are not autistic. If you see reduced joint attention, limited pretend play, atypical sensory responses, or unusual social interaction alongside the speech delay, request a developmental evaluation in addition to a speech evaluation.
Research on parent-mediated interventions suggests that embedded daily language-rich interaction (narration, shared reading, modelled recasts, responsive turn-taking) matters more than formal drill sessions. Five to fifteen minutes of quality parent-child interaction distributed through the day consistently outperforms one long drill session. Consistency matters more than intensity.
Children's speech output varies dramatically by context — a child may produce hundreds of words at home and ten at the SLP's office, or the reverse. This is normal and does not mean the therapy isn't working or that you are "seeing something that isn't there." Language samples from multiple contexts (home recordings, daycare observation, therapy transcripts) give a more accurate picture than any single setting. Share context-specific observations with your SLP.
This page is grounded in research on early language intervention, parent-mediated therapy, and developmental language disorder.
Additional reading: the American Speech-Language-Hearing Association (ASHA) for practitioner-level resources; the Hanen Centre for evidence-based parent coaching programs; Kaiser and Roberts on enhanced milieu teaching; Millar, Light, and Schlosser (2006) on AAC and speech outcomes.
This page is grounded in a formal Outcome-Driven Innovation (Ulwick, 2005) analysis of the unmet needs of parents of children with speech and language delays. ODI is a structured method for ranking desired outcomes by importance (how much does this outcome matter to the population?) and satisfaction (how well is the outcome currently served by existing solutions?). The opportunity score = Importance + max(Importance − Satisfaction, 0), scaled 1–20. Scores ≥ 15 indicate extremely underserved outcomes; 12–14.9 significantly underserved.
The speech delay parent analysis (completed 2026-04-03) harvested 30 desired outcomes from first-person parent quotes across speech therapy forums, SLP resources, app reviews, and the research literature. Outcomes were audited down to 27 validated ones and each scored on importance and satisfaction, then clustered into four opportunity areas.
| # | Outcome | Imp | Sat | Opp | Job step |
|---|---|---|---|---|---|
| 1 | Minimize the likelihood of failing to notice gradual improvements happening over weeks or months | 9 | 3 | 15 | Monitor |
| 2 | Minimize the likelihood of losing confidence in the therapy process due to invisible day-to-day progress | 9 | 3 | 15 | Monitor |
| 3 | Minimize the likelihood of dismissing early warning signs as "they'll grow out of it" | 9 | 4 | 14 | Define |
| 4 | Minimize the time it takes to detect whether the child is making meaningful progress | 9 | 4 | 14 | Monitor |
| 5 | Minimize the likelihood of skipping practice days due to burnout or busy schedules | 8 | 3 | 13 | Execute |
| 6 | Minimize the likelihood of inadvertently reinforcing speech-hindering habits | 8 | 3 | 13 | Prepare |
| 7 | Minimize the time it takes to learn effective speech-supporting techniques for daily use | 8 | 4 | 12 | Prepare |
| 8 | Minimize the time it takes to integrate speech exercises into daily routines without them feeling forced | 8 | 4 | 12 | Execute |
| 9 | Maximize the likelihood of having objective evidence of progress to share with professionals | 8 | 4 | 12 | Monitor |
| 10 | Maximize the likelihood of identifying which specific techniques are driving improvement | 7 | 2 | 12 | Modify |
Summary statistics: Average importance 7.5 / 10. Average satisfaction 3.9 / 10. Average opportunity score 10.9 / 20. Two outcomes score ≥ 15 (extremely underserved); six score 12–14.9 (significantly underserved). No existing parent-facing app combines progress tracking, practice consistency, technique education, and SLP session preparation.
1. Progress visibility over long timelines — avg opp score 14.5. Parents of children in speech therapy cannot see gradual language development over months or years. No existing tool provides a mechanism for detecting or surfacing incremental progress. Parents consistently abandon effective approaches, question the SLP, or drop out of therapy when slow improvement is actually occurring but remains invisible.
2. Daily practice consistency — avg opp score 12.7. Parents struggle to maintain consistent home practice across caregivers, busy weeks, and child refusal episodes without burning out or accumulating guilt. No mainstream app addresses practice-habit tracking specifically for speech support; generic habit trackers miss the embedded nature of language-rich interaction.
3. Parent skill building — avg opp score 12.3. Parents need to learn evidence-based techniques (Hanen, enhanced milieu teaching, recasting, expansion), avoid common speech-hindering habits (talking for the child, excessive screens), and identify which techniques are working. Information is scattered across books, YouTube channels, and SLP handouts, and contradictory advice is the norm.
4. Early recognition and action — avg opp score 12.5. Parents delay seeking help because they cannot distinguish normal variation from concerning delay, and "wait and see" remains a common default from pediatricians and family members. Milestone charts exist but are underused and often misread. Intervention delay is one of the strongest predictors of persistent difficulty.
The research on this page matters more than any app. Some parents find that a daily practice makes the frames easier to hold when the comparison tax lands or the practice streak breaks.
Other long-form research pages in the Unseen Progress library:
Unseen Progress. (2026). Speech and language delay research — the top 10 problems parents face during therapy. https://unseenprogress.com/research/speechstrong/