Your child is laughing at the tablet, tapping fast, clearly having a good time. Whether that is learning or just fun is genuinely hard to tell from the outside, and enjoyment tells you nothing either way. Plenty of real practice is enjoyable, and plenty of pleasant activity teaches nothing.
There are three things that separate them, and all three are observable from across the room.
Is anything being retrieved?
Watch what the child is being asked to do, not how engaged they look.
If they are tapping through screens, watching animations, matching things that are visible at the same moment, then nothing is being pulled out of memory. It is pleasant, and it is not practice.
If they have to supply something that is not on the screen, a word, a spelling, a meaning, then memory is doing work. That is the whole difference. A child who pauses, frowns, and then answers is learning. A child who never pauses is not being asked anything difficult.
The pause is the tell. If there are no pauses at all, the difficulty is set too low to matter.
Do mistakes lead anywhere?
The second thing to watch is what happens after a wrong answer.
In weak apps, a mistake costs a point or a heart, a red X appears, and the game continues to the next thing. The missed word may never return. The child learns that mistakes are a penalty to be avoided rather than information.
In something that works, a wrong answer changes what comes next. The word reappears, soon, and then again a few days later. Over a couple of weeks you should see the same troublesome words resurfacing rather than the app marching endlessly through new material.
That is worth checking directly: sit down for five minutes and see whether words that were missed earlier come back.
Does it get harder?
Difficulty that never moves is the clearest sign of a game that has stopped teaching.
If your child is answering nearly everything correctly and quickly, they have outgrown the level and are now just collecting points. If they are getting most things wrong, they are being battered and will quit.
The right zone is uncomfortable but survivable: most answers right, a steady minority wrong. If the app never adjusts, that zone lasts about a week and then the whole thing becomes a screensaver with rewards.
The test that settles it
None of this requires trusting a review or a marketing page.
Wait a few days after a session, then ask about five or six words in ordinary conversation, away from any screen. Point at something in the kitchen. Ask what a word means on a shop sign.
If the words come back outside the app, the play was doing its job. If they only exist inside the game, your child has been learning the game rather than the language, and the time would be better spent elsewhere.
It is a blunt test and a fair one. We would rather be judged on it than on how much fun the games look.
Related: the screen-time question
Try it with your class. GetSpeakin reads every answer and puts the words a child actually needs in front of them. Book a demo for your school.