Gujarati apps vs workbooks: what actually works for heritage families

Almost every Gujarati family we hear from started in the same place: typing learn to speak Gujarati app into Google, scrolling the app store results, downloading two or three, and quietly abandoning them a fortnight later.

That is not a discipline failure. It is a mismatch between what those apps are built to do and what a heritage learner actually needs. This is an honest attempt to explain the difference — including where our own workbooks are the wrong answer.

First, the thing nobody tells you: Duolingo does not teach Gujarati

Gujarati has somewhere in the region of 55 million speakers worldwide, which makes it one of the larger languages on earth — and it is still not on Duolingo. Neither are most South Asian languages beyond Hindi. If you went looking for the green owl and came away confused, that is why.

What does exist is a smaller field: Ling, LingQ, Indilingo, iVoca's listening-and-speaking app, and a handful of alphabet-and-vocabulary apps from smaller developers, plus tutor marketplaces like Preply and italki where you can book a human. Some of these are genuinely well made. None of them were designed with your family in mind, and that turns out to matter more than the quality of the software.

What apps are genuinely good at

Three things, and they are not small:

• Showing up. An app is in your pocket, it takes ninety seconds, and it nags. For building a daily habit, nothing on paper competes.
• Spaced repetition. Software is better than you are at knowing which word you are about to forget and putting it in front of you the day before you forget it.
• Hearing the language. Audio on demand, repeated as many times as you like without embarrassing yourself in front of a relative.

If you are trying to learn two hundred words, an app will get you there faster than a book will.

Where apps leave heritage learners stuck

1. They assume you are starting from zero. You are not.

This is the big one. Most diaspora adults are not beginners — they are passive speakers. You understand your parents fine. You follow the argument at the family lunch. What you cannot do is produce a sentence at speed under mild social pressure.

That is a specific, well-documented gap between comprehension and production, and almost no app is designed for it. They start you at hello, my name is, which you already know, and the early units feel patronising, so you quit before reaching anything that would have helped. The thing you actually need is structured practice at making sentences, not recognising them.

2. The vocabulary is tourist-shaped, not family-shaped

App courses for smaller languages are usually built from a standard template: greetings, numbers, the restaurant, the hotel, the train station, directions. Useful on a trip to Ahmedabad. Almost useless at your own dinner table.

What a heritage family needs is the opposite list — the language of the house. Have you eaten? Wash your hands. Put your shoes on. Sit properly. Who was on the phone? Tell Ba what you did at school. And above all the kinship vocabulary, because Gujarati is far more precise about relationships than English is: your father's older brother, your mother's sister, your paternal grandmother are all distinct words, and no app course front-loads them the way a family would.

3. Tapping is not writing, and the script needs a hand

Gujarati script hangs meaning on matras — the vowel marks that sit above, below and beside a consonant and change what it says. Children (and adults) understand that system far faster when they have drawn it than when they have selected it from four options on a screen. Handwriting is slow, and the slowness is the point: it forces you to notice the parts.

4. Streaks are not accountability

A 340-day streak is a relationship with an app. Progress in a heritage language comes from a relationship with a person who speaks it — and who will wait, patiently, while you assemble the sentence.

5. The content runs out

For a language the size of Gujarati, app courses are thin. Learners who stick with one often hit the end of the useful material in a few months and find nothing to progress to.

What workbooks are good at — and bad at

Paper wins on the things software cannot touch. It gives you handwriting practice. It lets a parent and child sit on the same side of the table looking at the same page, which is a different act entirely from two people on two devices. It provides sequence — someone has decided what comes after what — and it does not sit on a device that also contains YouTube.

And it is honestly worse at the rest. A workbook will not remind you it exists. It does not know which word you are about to forget. It cannot pronounce anything without an audio companion. Anyone who tells you paper beats software outright is selling paper.

The combination that actually works

The families who get somewhere almost all end up with the same three-part stack, and the third part is the one people skip:

1. Daily repetition, five to ten minutes. An app, or flashcards, in the supermarket queue and the school pick-up line. This maintains vocabulary.
2. Structured study, once or twice a week, thirty to forty minutes. A workbook, a class, or a tutor. This builds the script, the grammar and the sentence patterns. Repetition alone never produces these.
3. Real conversation with a real person, once a week, non-negotiable. A video call with a grandparent. Ten minutes in Gujarati at Sunday lunch, however badly it goes. This is the part that converts study into speech, and it is the part everyone postpones until they feel ready. You do not become ready first. You become ready by doing it badly for a while.

If you only have the appetite for one of the three, choose the third. A learner with weekly conversation and no materials will outrun a learner with perfect materials and no one to talk to, every time.

A realistic weekly rhythm

For a family with school-age children, something like this survives contact with real life:

• Monday to Friday: ten minutes of flashcards or app practice, attached to something that already happens — after dinner, before a bath, in the car.
• One weekday evening: twenty to forty minutes of workbook. Children do tracing and words; the adult does their own lesson at the same table. Co-learning beats supervision — children work much harder when a parent is visibly also struggling.
• Sunday: the phone call. Same time each week, so it becomes a fixture rather than a plan.

That is roughly ninety minutes a week. It is unglamorous and it compounds.

So where do our products sit in this?

Honestly: we make part two, and a piece of part one. Our children's workbook and adult course are the structured-study layer, with audio for pronunciation; the 280 flashcards handle daily repetition away from a screen. The Family Learning Bundle is the whole stack for a household, which is how most families actually use it.

We do not make part three, and neither does anyone else. That one is your family.

If there is a Gujarati class near you, take it — a Sunday school plus a weekly call will beat anything you buy. We have listed the ones we know of in learning Gujarati in New Zealand.

Common questions

Is there a Gujarati course on Duolingo?
No. There is no Gujarati course on Duolingo. The main app options are Ling, LingQ, Indilingo and several smaller alphabet and vocabulary apps, plus tutor marketplaces for one-to-one lessons.

Can an app alone make me conversational?
It can build vocabulary and listening. It will not reliably produce speech, because speaking is a separate skill that only develops under the mild pressure of a real conversation.

My child refuses the workbook but likes the app. Is that fine?
For now, yes — anything they do willingly beats something they resist. But keep the handwriting alive in small doses; it is the part that does not come back easily later.

Should I learn the script or stay in roman letters?
Learn the script. Roman transliteration cannot represent several sounds Gujarati distinguishes, so learners who stay in roman letters develop errors they cannot hear. It costs a few evenings and removes a permanent ceiling.

What if I am not fluent enough to teach my own child?
That is the normal case, not the exception — and it is fixable. See how to teach your child Gujarati when you are not fluent yourself.