How do I know if a speech app is actually working?
Judge it on behaviour outside the app, not on anything it reports. Six signs arrive roughly in order: your child asks for the app, imitates words inside it, fills the pause before it prompts, imitates a word outside it, produces longer attempts, and finally uses a word to ask for something. Sign four - imitation outside the app - is the one that matters, because it is the only evidence the practice transferred. Pick three target words, tally spontaneous uses for four weeks, and if nothing moves on any sign, stop relying on the app and see a speech-language pathologist.
Three weeks in, a fair question arrives: is this doing anything, or is my two-year-old just enjoying a nicely made toy?
It deserves a real answer, and the honest one starts with what not to look at. Our app does not score your child, and no app should pretend to — measuring speech progress properly means standardised assessment by a clinician. What you can do is watch for six specific things at home, in a sequence that tends to hold, and know which one actually counts.
What Are the Six Signs a Speech App Is Working?
They usually arrive in this order. Seeing an early one is real progress even when the later ones have not shown up yet.
| # | Sign | Typically | What it tells you |
|---|---|---|---|
| 1 | They ask for it by name — or by vehicle | Week 1 | Engagement. Not progress yet, but the precondition for all of it: a child who refuses the app cannot practise with it. |
| 2 | They imitate words inside the app | Weeks 1–2 | Imitation in a supported, predictable routine. The easiest form of production, and the first real skill. |
| 3 | They fill the pause before the prompt | Weeks 2–3 | Anticipation. They have learned the pattern and are now producing on their own timing rather than following. |
| 4 | They use a word outside the app | Weeks 3–4 | The one that matters. Transfer to real life is the only sign that shows the practice went anywhere. |
| 5 | Their attempts get longer, not just clearer | Weeks 3–6 | Growing capacity. "Ba" becoming "buh-ka" for bucket is progress even though it is still not the word. |
| 6 | They use a word to get something | Weeks 4+ | Functional use. Saying "more" to be given more is different from saying "more" because you asked them to. |
The row numbered 3 is the one parents most often miss, because it looks like nothing: the app pauses, and your child speaks into the pause without waiting to be asked. That half-second is the difference between following a prompt and choosing to talk.
Why Does Sign Four Matter More Than the Others?
Because everything before it happens inside an environment built to make it happen. The app knows which word is coming, it pauses at exactly the right moment, the picture is on screen, and the same routine ran yesterday. That scaffolding is the point — it is what makes practice possible for a two-year-old. But it also means success inside the app proves only that the scaffolding works.
A word said in the kitchen, about a real truck, with nobody prompting, is a different claim entirely. That is your child using the word, not completing the activity. In speech-language terms this is generalisation, and it is what any intervention is ultimately judged on. If you get one instance of it in four weeks, the practice is going somewhere.
What Is Not Evidence of Progress?
Three things feel like proof and are not. Worth naming, because they are the metrics an app is most tempted to show you.
- Minutes used. Time on task is an input, not an outcome. Forty minutes of tapping through favourite screens can contain fewer speech attempts than eight minutes of real practice.
- Levels or activities completed. Completion measures navigation. A toddler can finish everything by tapping, and many will.
- "She loves it." Genuinely good news, and sign one — but a child can love an app for months without it changing their speech. Enjoyment keeps the door open; it is not what walks through.
This is also why Talk & Listen does not hand you a progress score. We could generate one. It would be a number about our own activities, dressed up as a claim about your child, and a reassuring number is exactly the thing that delays a referral worth making. What the app is for is a lot of cheerful repetitions; measurement is a clinician's job.
How Do You Track It Without Turning It Into a Chore?
Pick three target words. Not ten. Three words your child is close to but not reliable on, and ideally ones they actually want — "more", "car", "up", or whichever vehicle currently runs their life. Write them on the fridge.
Then, for four weeks, make a mark whenever your child says one of them without being asked. That last part is the whole method. Prompted repetition — you say "say truck", they say "truck" — is sign two, and you will already know whether it is happening. The tally is for spontaneous use, which is easy to miss and easy to forget by evening.
A fridge and a pen genuinely beats an app for this, because you will actually do it. If you would rather have something printed, the early sounds checklist gives you a sheet to work from, and the late talker milestones guide will tell you which words are reasonable targets for your child's age.
What Should You Do at the Four-Week Review?
Look at the tally and the six signs, and pick one of three answers.
More marks than week one
It is working. Keep the same three words until they are solid, then swap in three more. Do not add words because progress feels slow — three at a time is what makes the progress visible.
Signs 1–3 but nothing outside the app
Practice is landing but not transferring. Take the words off the screen: same word, real toy, pause where the app pauses, then wait longer than is comfortable. Most of us fill the silence in under two seconds.
No movement on any sign
Stop treating the app as the plan and book a speech-language pathologist. Four weeks of consistent practice with nothing to show is information, and it is the kind worth acting on early.
When Should You Call a Speech-Language Pathologist Regardless?
Some things are not app questions, and no amount of practice at home replaces an assessment. Book one — without waiting out a four-week experiment — if any of these are true:
- Your child is over 30 months with fewer than 50 words, or no two-word combinations
- They have lost words or skills they previously had
- Comprehension seems affected too — following simple instructions is hard, not just speaking
- There are feeding or swallowing difficulties alongside the speech
- Your own instinct says something is off. Parents are usually right about this, and early is cheap
An evaluation is not a commitment to therapy. It is information, and it is the one thing that tells you whether four more weeks of practice is the right plan or a delay. If you do start therapy, keep the app if your SLP is happy with it — home practice between sessions is one of the better predictors of progress, and the usual reason practice stops is that it became a chore rather than that it was the wrong activity.
The Short Version
Watch what happens away from the screen. Three words, four weeks, spontaneous uses only. One word said unprompted in real life beats a month of perfect in-app performance, and four weeks with nothing is a reason to get a professional opinion rather than to try harder.
And when the answer turns out to be yes — words arriving, tally climbing — the next question follows immediately: what to practise next. Short answer: not more nouns.
Ad-free & offline ready
